Agenda packet — Modesto City Council, June 4, 2024 (part 7 of 11)

SAP HANA Enterprise Cloud administrative tasks will be done using management networks. Administrative access to the management networks from the SAP internal networks is only possible using dedicated jump hosts with strong authentication.

Security of the SAP internal network including SAP end user equipment is ensured using solutions like network admission control, Intrusion Prevention Systems, network filtering, strong authentication for remote access, internet content filtering, anti-virus scanner.

SuccessFactors We serve our customers and end users from secure data centers around the world. Physical security features at these facilities include a 24x7x365 manned security station and biometric and man-trap access controls. The systems at these facilities are protected by firewalls and encryption technology. Operational redundancy features include redundant power, onsite backup generators, and environmental controls and monitoring.

We employ a wide range of security features, including two factor authentication, data encryption, encoded session identifications and passwords. Our hosting providers conduct regular security audits of our infrastructure. We also employ outside vendors for 24x7x365 managed network security and monitoring. Every page we serve is delivered encrypted to the end user via a Transport Layer Security or TLS. We also use encryption technology in our storage systems.

We continuously monitor the performance of our cloud offerings using a variety of automated tools. The architecture is designed with built-in redundancy for key components. We load balance at each tier in the network infrastructure. We also designed our application server clusters so that #15..
- 826 -
SOLICITATION # CH16012
88 servers can fail without interrupting the user experience, and our database servers are clustered for failover. We regularly back up customer data.

We have implemented a multi-tiered architecture leveraging a strategy of “defense in depth” with 6 tiers of virtual networks (VLAN) for separation at each delivery layer. Network traffic is logged and monitored with live monitoring through an Intrusion Detection System (NIDS), and controlled through a series of switches and routers whereas data must pass through each tier in order to get to the next tier.

In addition to Physical (site) security, the logical network stratification includes the following:

Tier 1 (External VLAN/firewall) - The first tier consists of the external network and perimeter firewalls. These provide an initial layer of defense and protect the following layers from unauthorized access. Note that port 443 (HTTPS for web traffic) is the only port open.

Tier 2 (Internal VLAN/firewall) - A de-militarized zone (DMZ) exists with load balancers. The DMZ provides a second line of defense while the load balancers are the first layer of scalability for the service delivery. The DMZ functions as a ""neutral zone"" between the network and the outside public network.

Tier 3 (Web VLAN) - The Web tier presents the User Interface to the application, and separates the application, reporting and utility servers from the other tiers.

Tier 4 (Application VLAN) - The Application tier contains the business logic and transaction servers, and is managed through clustered, high availability (HA) servers. Pre-configured as “Pods”, additional servers can be added as needed to provide scalability and performance.

Tier 5 (Database VLAN) - The database tier is protected by an additional set of perimeter firewalls. The database processing is executed on database servers leveraging a multi-tenant, fully qualified database schema.

Tier 6 (Storage VLAN) - Data is persisted to disks with include a Storage Area Network (SAN). Prior to store, data is encrypted by way of data appliance with AES-256 bit encryption.

VMware VMware IaaS Services
At the tenant level firewall services are provided via the vCloud Networking & Security Edge Gateways for customers to configure and maintain. Firewall policies can be used to restrict and manage public/Internet-based traffic and create DMZ zones for multi-tier applications. Firewall policies can also be used to configure access policies between internal IP networks and VxLAN segments. Stateful inspection firewalling can be applied on the external interface of the vCNS Edge Gateway.

VMware architects, provisions, monitors and manages the vCloud Hybrid Service infrastructure and surrounding components. As described in the AT 101 (ISO 27001) report, access points such as delivery and loading areas and other points where unauthorized persons may enter the premises are controlled and, if possible, isolated from information processing facilities to avoid unauthorized access. The details of this control are not disclosed publicly.

VMware AirWatch
We implement multiple security measures to protect hosted servers, including physical and logical controls. Due to FOIA requirements and the competitive EMM marketplace, we cannot provide specific architecture details at this time.
#15..
- 827 -
SOLICITATION # CH16012
89 FireEye High-level architecture diagrams following industry best practices utilizing both virtual and physical controls are available and contained in the SOC 2 report.

VirtueStream Virtustream utilized a virtual local area network (VLAN) and virtual route forwarding (VRF) to logically segregate all customers in the IaaS environment. Every customer chooses their own IP Address range, as there is no ability to overlap IP Addresses between customers. In addition, every customer is logically separated with Firewall Context with their own rule, where by default everything is denied except specific rule.

8.6.10 Provide Security Technical Reference Architectures that support Infrastructure as a Service (IaaS), Software as a Service (SaaS) & Platform as a Service (PaaS).

CA APM Only SaaS is provided, network diagram can be provided upon signing NDA MAA Only SaaS is provided, network diagram can be provided upon signing NDA CA Agile These can be provided when a NDA is in place.

ASM Only SaaS is provided, network diagram can be provided upon signing NDA Salesforce Salesforce offers market leading PaaS and market leading SaaS solutions. Salesforce does not provide IaaS as a direct service offering to our customers, it is an underlying part of our PaaS and SaaS offerings.

The Salesforce Platform is built for cloud computing, with multitenancy inherent in its design. To meet the high demands of its large user population, Force.com’s foundation is a metadata-driven software architecture that enables multi-tenant applications.

Force.com combines several different persistence technologies, including a custom-designed relational database schema, which are innately designed for clouds and multitenancy—no virtualization required.

Force.com’s core technology uses a runtime engine that materializes all application data from metadata—data about the data itself. In Force.com’s well-defined metadata-driven architecture, there is a clear separation of the compiled runtime database engine (kernel), tenant data, and the metadata that describes each application. These distinct boundaries make it possible to independently update the system kernel and tenant-specific applications and schemas, with virtually no risk of one affecting the others.

Every logical database object that Force.com exposes is internally managed using metadata.

Objects, (tables in traditional relational database parlance), fields, stored procedures, and database triggers are all abstract constructs that exist merely as metadata in Force.com’s Universal Data Dictionary (UDD). For example, when you define a new application object or write some procedural code, Force.com does not create an actual table in a database or compile any code. Instead, Force.com simply stores metadata that the system’s engine can use to generate the virtual application components at runtime. When you need to modify or customize something about the application schema, like modify an existing field in an object, all that’s required is a simple nonblocking update to the corresponding metadata.

Because metadata is a key ingredient of Force.com applications, the system’s runtime engine must optimize access to metadata; otherwise, frequent metadata access would prevent the service from scaling. With this potential bottleneck in mind, Force.com uses massive and sophisticated metadata caches to maintain the most recently used metadata in memory, avoid performance-sapping disk I/O and code recompilations, and improve application response times.

#15..
- 828 -
SOLICITATION # CH16012
90 The multitenant architecture and secure logical controls address separation of Customer Data. The Salesforce infrastructure is divided into a modular architecture based on “instances”. Each instance is capable of supporting several thousand customers in a secure and efficient manner. Salesforce uses the instance architecture to continue to scale and meet the demands of our customers. There are appropriate controls in place designed to prevent any given customer’s Salesforce instance from being compromised. This functionality has been designed and undergoes robust testing through an on-going process by both Salesforce and its customers.

QTS The QTS cloud makes use of unique managed service provider architecture layer(s). Information systems, particularly those based on cloud architecture models, are made up of different service layers. The layers of the QTS cloud that are defined in its System Security Plan, and are not leveraged by any other Provisional Authorizations, are: Infrastructure as a Service (IaaS). Note:

Please refer to NIST SP 800-145 for information on cloud computing architecture models.

SAP Ariba Ariba recognizes that security is a critical component of effective electronic commerce architecture and takes necessary security measures to protect any information passed between buyers and suppliers. Ariba implements security using a variety of hardware, software, and procedural best practices. Full details about our security mechanisms and procedures are available in Chapter 3 of the Ariba Cloud Technical Infrastructure Whitepaper.

Hanna N/A, HANA Enterprise Cloud is a private managed cloud SuccessFactors We understand the critical importance of information protection and recognize the contribution that information security makes to an organization’s strategic initiatives and overall risk management.

We have implemented security controls and practices designed to protect the confidentiality, integrity, and availability of customer information. We continually work to strengthen and improve those security controls and practices as well.

The current best practices associated with information security involve a layered approach, what the industry calls “defense in depth.” Regardless of the software delivery model, security cannot be implemented at a single “make or break” point. For a SaaS provider to facilitate data security for sensitive information, it must have a comprehensive, multifaceted security program in place. We take a holistic approach to information security, implementing a multilayered defense at all the touch points in the information flow—both the physical and logical, applied across the database, middleware, application, and network and communication layers—to offer complete data privacy, transparency, and audit controls.

VMware Compliance Reference Architecture Framework (RAF) that provides a consistent method for VMware, its partners, and customers to assess and evaluate the impact of regulations on virtual and cloud environments. The intent of the RAF is to provide a single framework for VMware, its partners, and organizations to address a variety of compliance requirements across an IT infrastructure. This includes: Product Applicability Guide (PAG), Architecture Design Guide (ADG), and a Validated Reference Architecture (VRA).

FireEye High-level architecture diagrams following industry best practices utilizing both virtual and physical controls are available and contained in the SOC 2 Type 2 report.

VirtueStream Virtustream Cloud Platform Security (for the Infrastructure-as-a-Service environment) is designed, built, and operated to provide highest level of infrastructure security available.

Proposed State of Utah Security Architecture
#15..
- 829 -
SOLICITATION # CH16012
91 Figure 1. xStream Enterprise Architecture
State of Utah end users would access the hosted SAP environments via a MPLS connection (Provided by State of Utah). Additionally a VPN connection has been sized at 100Mbps however this can be decreased or increased depending on the exact requirements.

State of Utah end users accessing our enterprise platform would first hit the landing zone in a dedicated virtual local area network (VLAN) and dedicated virtual route forwarding (VRF). The next hop into the enterprise platform is a dedicated firewall (FW) context on a Cisco Firewall services module. After traversing the core, traffic hits another dedicated VRF and drops into the platform network and compute (CPU and Memory essentially a blade server) layer. All traffic is VLAN separated. At each compute host, a hypervisor based firewall and intrusion protection system (IPS) provides a dedicated client rule set to further ensure network security. All traffic transitioning from one zone to another are monitored by Intrusion Detection and Intrusion Prevention systems.

Traffic coming in via the internet hits the front-end perimeter with load balancing modules and multicontext Cisco firewalls. All traffic then passes through intrusion detection system and intrusion protection systems. As in the enterprise compute layer, every host in our demilitarized zone (DMZ) has a hypervisor-based firewall and IPS with dedicated client rule sets. Again, all traffic is VLAN separated as well.

Traffic that needs to traverse from DMZ to the enterprise runs through another dedicated FW context and IDS/IPS.

All of the State of Utah environments and data will be hosted in Virtustream’s data centers within the continental United States.

Standard Services used in Virtustream’s management environment and in all client environments including the following:
#15..
- 830 -
SOLICITATION # CH16012
92  All Virtustream employees use Mandatory Strong 2 factor authentication (2FA OTP) Administrative Access to all systems.
 Dedicated VLAN network segmentation and dedicated Virtual Route Forwarding (VRF) are used extensively to segregate environments and zones.
 Perimeter Firewalls are used to segment internal and external environments as well as segregate security zones. Configuration, monitoring, auditing and logging are included.  Virtual Machine-based Firewall and Intrusion Prevention System (IDS) is installed on every virtual machine in the environment and is protected with Juniper’s Security Gateway virtual firewall application and monitored service.

Security Services that are standard components for Virtustream’s management environment and are Optional Services for client environments including the following:
 Managed Two-Factor Authentication (“2FA”) is in use for all application systems.  Intel TxT Enabled Servers and Trusted Boot/Bios monitoring with Attestation Server and OS and VM support, including Geolocation and Geofencing according to NIST 7904 guidelines.  Secure operating system (OS) builds based on DoD Secure Technology Implementation Guide (STIG) guidelines are used to build Virtustream’s Management and Administration Servers.  Server/File Integrity Monitoring (FIM) is installed in the PCI and VFC clouds.  Patching Regimen: Virtustream patches host servers, network devices, security devices, servers and related services in the Management Network on a specified routine (monthly or quarterly, depending on release schedules), or when there is a CERT or other authorized source of patch that requires immediate attention. Based on urgency and risk of the issue we will schedule the patch as appropriate and use change control.  Scanning regimen: vulnerability scanning is done on a monthly basis with additional 3rd party vulnerability scans done monthly. Additional scans are done when made aware of new vulnerabilities. Issues are classified and addressed according to Risk Classifications and are addressed with ITIL v3 change control processes.
 Managed IDS signatures are routinely updated and the logs are monitored.  Anti-Virus is managed on all Management servers in Enterprise, PCI and, VFC clouds.  Logging Service of all servers, network devices, and security devices to a centralized log server system.
 Governance Risk Compliance: We use a complete Enterprise Risk Management toolset to manage compliance reporting and continuous monitoring to all of our supported compliance frameworks.

Virtustream’s facility monitoring systems are complete as per specifications in NIST 800-53r3/4. We use site assessment methodologies and checklists as detailed in NIST 800-42. Our systems and facilities are monitored 24/7 for any exceptions or trends. Our tools, processes and CONUS (Continental USA) personnel monitor network, power, cooling, humidity, water leakage, fire suppression, power systems (utility power, UPS systems and generators) and site access.

Virtustream’s Physical Access Control Security is designed to protect the confidentiality, integrity, and availability (“CIA”) of the cloud platform system and its data with the following security components:
 Limited and controlled room access.
 Logged and monitored access of all access control events.  Video surveillance and review of all access control events.  Biometric access control required to gain access to the Data Center.  US Data Center staff is limited only to US Citizens.
#15..
- 831 -
SOLICITATION # CH16012
93  Locked racks and rooms with key log out/in process.  Data destruction policies and procedures.
 Asset in/out policy and procedures.

Physical access by authorized staff is controlled by badge systems and biometric access systems.

All access of any kind is recorded and logged. Access lists are reviewed every 90 days. Employees that do not have a reason to have physical access are removed from our access management systems. Multiple high resolution and IR enhanced CCTV Cameras monitor our datacenters at all egress and ingress points as well as other sensitive areas. Security video footage is stored in a secure area for 90 days.

Approved Visitor Access requires VISITOR Badges and bright lanyards that are specifically different from employee badges. Visitors are required to produce and surrender a state or government issued photo ID during the visit, have a pre-approved reason for the visit, and have the escort witness the sign in and out procedure. Visitors are escorted continuously through approved areas of the facility.

Visitors are not allowed access to following areas of the data center under any circumstances:
 Cloud hosting areas.
 Networking and telecommunication areas.
 Guard areas.
 Power vaults.

8.6.11 Describe security procedures (background checks, foot printing logging, etc.) which are in place regarding Offeror’s employees who have access to sensitive data.

FireEye Background verification for employment candidates is a mandatory component of FireEye’s hiring process. All personnel are required to sign a confidentiality and non-disclosure agreement agreeing not to disclose proprietary or confidential information including client information to unauthorized parties.

Security awareness training program is in place to maintain the skill level of personnel regarding security and privacy expectations and best practices. All FireEye personnel at all levels are trained and notified of information security and privacy requirements and personnel responsibilities.

VirtueStream Virtustream employees who are assigned to IaaS must pass a Virtustream background investigation. In addition, Virtustream employees assigned to the IaaS must adhere to any requirement by customers to pass federal, state, or local background investigations if they are to provide managed services to the customer zone which includes access to sensitive data.

8.6.12 Describe the security measures and standards (i.e. NIST) which the Offeror has in place to secure the confidentiality of data at rest and in transit.

CA At CA Technologies, we comply with a multifaceted security standard that includes requirements for security management, policies, procedures, network architecture, software design and other critical protective measures. We ensure the maintenance of vulnerability management programs, implement strong access control measures, regularly monitor and test networks and adhere to the highest industry compliance and security policies. CA offers a variety of SaaS solutions, details for each offering has been provided in Exhibit 1 and 2 of this proposal.

Google All connections from customer end point devices to Google's Front End Servers are encrypted with enforced HTTPS sessions using Forward Secrecy. Google websites and properties use robust public key technologies: 2048-bit RSA or P-256 ECDSA SSL certificates issued by a trusted authority (currently the Google Internet Authority G2). #15..
- 832 -
SOLICITATION # CH16012
94 All data in transit between Google's Data Centers traverses across Google's private fiber network using a customized, proprietary encryption technology.

Google hard drives leverage technologies like FDE (full disk encryption) and drive locking, to protect data at rest.

These methods of encryption are fully managed by Google and Google's Key Management Servers based on 128-bit or stronger Advanced Encryption Standard (AES).

Encryption Keys and Ciphers Supported by Google
Protocols
TLS 1.2
TLS 1.1
TLS 1.0
SSL 3.04
QUIC
Cipher suites
ECDHE_RSA with AES
ECDHE_RSA with 3DES
ECDHE_ECDSA
RSA with AES
RSA with 3DES
Signing keys
RSA 2048
ECDSA P-256
Hash functions
SHA384
SHA256
SHA1
MD5
Salesforce Government Cloud Encryption Capabilities:

As part of the Salesforce Government Cloud, Salesforce is capable of responding to FIPS 140-2 cryptographic implementations for data being transferred between the State’s web browser and Salesforce. Data that resides within Salesforce’s protected boundary does not use FIPS 140-2 validated encryption as compensating/mitigating controls are in place to protect data. Additional information is provided below.

Data In Motion:

Salesforce employs cryptographic mechanisms to protect information during transmission. All transmissions between the user and Salesforce are encrypted by default with a 2048- bit Public Key. Our service uses International/Global Step Up certificates. We support one- way TLS, in which customers create secure connections before sharing private data.

Secure routing and traffic flow policies ensure that customer traffic is encrypted entering Salesforce until the load balancer decrypts the traffic. The load balancers decrypting the traffic are FIPS 140-2 compliant and are located inside of the Salesforce Government Cloud isolation boundary.

Data At Rest:

NIST SP 800-53 Rev. 4 states in SC-28, “Information at rest refers to the state of information when it is located on a secondary storage device (e.g., disk drive, tape drive) within an organizational information system.” SC-28 also states, “Organizations may choose to employ different #15..
- 833 -
SOLICITATION # CH16012
95 mechanisms to achieve confidentiality and integrity protections, as appropriate.” All secondary storage media (hard drives, disk drives, and tapes) containing customer data are maintained within Salesforce’s secure production data centers until the media has been sanitized and destroyed. Salesforce relies on physical access controls as a compensating control to protect the data.

For primary data storage, Salesforce provides customers with a built-in capability to apply fieldlevel encryption, using 128-bit keys with AES encryption, for a selection of custom fields included in the Force.com Platform and Salesforce Services. Field-level encryption ensures the data associated with designated fields is encrypted in storage.

SAP Ariba Sensitive data elements, including PCI Primary Account Numbers and system account parameters for internal application communications, are stored in Ariba databases encrypted by the AES (Advanced Encryption Standard) and Triple DES (Data Encryption Standard) encryption algorithms with a minimum key length of 128 bits. Encryption technology is also applied for the client connection to the Web site and to the hosted application passwords in storage. Customer user passwords are one way hashed using SHA256 and salted with random data. Limited Ariba Operations personnel have data query access and monitoring rights for the Ariba Hosting program.

Fieldglass
Hanna SAP provides for enterprise-grade and industry-standard security. SAP HANA Enterprise Cloud datacenters are enterprise-class security with enterpriseclass protection including data encryption, network encryption, firewalling, network isolation, and intrusion detection. HANA data is protected in multiple different ways. SAP HANA Enterprise Cloud is a managed cloud service that employs either a VPN or MPLS connection for transit. By default VPN is encrypted and MPLS can be encrypted. The customer procures the method of connection. At installation, HANA data can be encrypted using a feature called Data Volume Encryption. This encryption protects data in the persistence layer. Storage volume access is restricted to the customer account that created the volume, thus denying all other customer accounts the permission to view or access the volume that includes data isolation, masking, zoning and Logical Unit (LUN) binding. Strict user and access management, authorization management according to the need-to-know principle for administrative accounts, and security logging and security monitoring for critical activities or access, also protect data stored in HANA.

Hybris The Hybris platform uses SSL (https) for both the web application tier (browsers and WebService APIs) as well as communicating with back office systems.

Data at rest is something that has to be taken into consideration in the application requirements by the partner/customer or PS teams. If encryption is required the application teams need to work on implementing the methods to encrypt and secure the data, for example encrypting data fields in the database.

SuccessFactors Each customer’s data is maintained in a separate database schema eliminating data segmentation breaches. Each schema has separate authentication credentials and assigned resource profile to restrict access rights and resource consumption. Encryption for data at rest uses the AES256 cipher. All application access is encrypted-in-transit over HTTPS with 128-bit TLS encryption.
#15..
- 834 -
SOLICITATION # CH16012
96 FireEye At rest: Data are collected using automated batch and real-time processes and by manual personnel-driven actions via an Analyst facing Portal. Any information that is collected may be stored temporarily within appliances attached to our customers’ networks and may subsequently be transmitted back to one or more FireEye datacenters. Data transmitted from a customer’s network to a datacenter travels over a strong encryption Virtual Private Network (“VPN”) established by the appliances.

At transit: FireEye’s network sensors are configured to collect full packet capture of customer network streams, including those originating from or destined to potentially malicious IP addresses, matching signature-based network indicators of compromise, results of domain name lookups, and full e-mail messages including headers, content and attachments. This information is primarily stored on-site on the appliances and only meta data matching FireEye / Mandiant indicators of compromise are sent to FaaS.

VirtueStream Virtustream’s use of Intel® Trusted Execution Technology (Intel® TXT), a hardware-root of trust security feature, and IntelÆ AES-NI encryption acceleration helps secure and protect virtualized environments against malware and provides more of the critical infrastructure and data protection assurances needed to enable trusted multi-tenancy in the cloud. Virtustream offers solution for data-at-rest and data-in-transit encryption solution for our customers as optional service, as most of our customer environment is only accessible via private network and only to the internal users of the customer. Solutions available to our customer as follows.

1. IPSEC VPN Tunnel for data-in-transit and

2. VM level encryption for files and databases

3. Encrypted Backup (all backups are encrypted by default) The technologies of encryption available throughout the entire data lifecycle and all are FIPS 140- 2 compliant.
a) DB and File System Encryption
b) Encrypts file system and volume data transparently to:
i) Applications
ii) Databases
iii) Storage Infrastructure
c) Integrated Key Management
d) High Efficiency Encryption
i) Centralized Key Management
ii) Policy Management
iii) Detailed Auditing
iv) Highly Available
v) FIPS 140-2 Certified Hardware Appliance
vi) Very Low Encryption Penalty due to Intel AES-NI enabled chip sets used e) Encryption of Virtual Machine, OS and App
i) Integration of SafeNET Protect-V
ii) Logging of all image access
f) Encryption in Archive
g) Implements Encryption, Access Control, Auditing on Host (LUW) h) Kernel Level Driver – Win
i) File System – Unix/Linux
j) Support for file systems and raw partitions
k) Highly efficient block encryption

8.6.13 Describe policies and procedures regarding notification to both the State and the Cardholders of a data breach, as defined in this RFP, and the mitigation of such a breach. #15..
- 835 -
SOLICITATION # CH16012
97 CA CA Technologies abides to contractual requirements for notification, in addition to working with legal to ensure compliance with regulatory requirements. There may be cases where incident analysis completion is a requirement for confirming the breach, and / or its impact and data leakage boundary. Prior to completion of this activity, CA may not have the needed conclusions for customer communication. Once subscriber data has been identified as part of the investigation, said subscribers will be notified as soon as possible and not longer than 5 days. A report of the incident will be available and distributed to clients within 30 days.

Please note that none of the offerings proposed as considered payment products and are not intended for use of storage of cardholder data.

Google Google will take and implement appropriate technical and organizational measures to protect Customer Data against accidental or unlawful destruction or accidental loss or alteration or unauthorized disclosure or access or other unauthorized processing. If Google becomes aware of a Data Incident, Google will promptly notify Customer as permitted by law of the Data Incident, and take reasonable steps to minimize harm and secure Customer Data. Notification(s) of any Data Incident(s) will be delivered to the Notification Email Address provided by the Purchasing Entity in connection with the Agreement or, at Google’s discretion, by direct communication (e.g., by phone call or an in-person meeting).

Salesforce Salesforce will promptly notify the State (within 48 hours, reflected in a negotiated agreement between the parties) in the event Salesforce becomes aware of an actual or reasonably suspected unauthorized disclosure of Customer Data. Notification may include phone contact by Salesforce support, email to the State's administrator and Security Contact (if contact is submitted by State and contact information is kept up to date), and public posting on trust.salesforce.com.

SAP Ariba Ariba has developed a computer security incident response team. Following policies and procedures, this team responds to suspected security incidents to mitigate risks and damage. The team conducts response and forensic analysis of systems and network traffic. This information can be used to assist with prosecution if a security breach is detected and the offender caught.

Fieldglass SAP Fieldglass’ security team is responsible for managing security incidents and all communication is conducted via the respective account managers to ensure timeliness. The process is defined within the Incident Response Management Standard.

Customers are notified of an incident within 48 hours.

Hanna SAP will notify via defined communication channels within 36 hours of a confirmed data security breaches to the affected customers. The report will detail the following information:
• Details relating to the security incident that has occurred, known at the time of notification.
• IT infrastructure and/or application affected by the security incident. • Overview of the performed mitigation actions to restore the security, documented within the incident report form.
• All further applicable requirements by country regulations “on obligation to notify” will be met.

Hybris Since we do not allow for CC data to be transmitted, stored or processed in our environment, we regularly scan the customer environment in case this is done without our consent.

However, we do have a CSIRP in place, in which we would notify the customer and our internal Forensics team in the event that a customer environment should be breached.

SuccessFactors Upon the occurrence of a data breach, initial communication is distributed to the appropriate individuals and an escalation process is followed. Upon #15..
- 836 -
SOLICITATION # CH16012
98 becoming aware of the incident, measures are promptly taken by the team to resolve the situation. All affected customers should be informed within at most 36 hours of confirming a potential breach in the privacy of their data. Following incident resolution, follow-up is required to ensure that the incident has been resolved effectively and that the threat is no longer present. We are aligned with ISO 27k standards for event and incident management and have formal incident management policies and processes in place. These policies and procedures are tested in the ISO 27k and SOC 2 audits.

FireEye FireEye has a documented policy for incident management that has been approved by management and communicated to appropriate constituents and owners. It is continuously maintain and reviewed annually. The plan has a reporting structure and escalation path. An incident response team with defined roles and response related qualifications are available 24x7x365. The team maintains chain of custody for evidence during the incident investigation.

There is a process for reporting client notification for regulatory, legal and contractual issues after a breach if a breach were to be confirmed. If a confirmed breach were to occur a management team would review all the factors and develop a remediation plan to mitigate.

VirtueStream There are two primary Incident types; Security Incidents, where there is a possible breach in systems or data integrity, and Services, where there is in impacted or affected service. Although there is some overlap, generally any security-related incident should be classified as a Security Incident and the response must be managed using the ‘Virtustream Information Security Procedure – Security Incident Response Plan’.

The purpose of this security incident response plan is to provide general guidance to Virtustream staff- both technical and managerial – to enable quick and efficient recovery from security incidents; respond in a systematic manner to incidents and carry out all necessary steps to correctly handle an incident; prevent or minimize disruption of critical computing services; and minimize loss or theft of sensitive or mission critical information including sensitive credit card data. The plan is also a guide to sharing information with other stakeholder organizations who might be impacted by such security incidents such as the credit card associations and law enforcement.

The Security Incident Response Plan (SIRP) provides guidance to prepare for, respond to, and recover from potential incidents. Policy statements surrounding the IR-Plan are provided to ensure continued upkeep and standardized use. The SIRP guidance at the procedural level defines the roles, responsibilities, communication methods and flows, contact information, types of potential incidents, and immediate actions that are to be taken upon an incident’s identification, and elaborates with subsequent recovery steps. Virtustream’s Incident Response Policy requires the implementation and testing of a generalized plan that adheres to the International Standards Organization 27002 guidance for incident management and response, but meets specific requirements for compliance such as PCI-DSS.

The Plan covers the corporate environment associated with Virtustream’s IT assets, the local IT resources and the IT resources at Virtustream’s Data Center. It consists of a series of guidelines (Incident Response Guidelines or “IRG”) that should generally be followed as appropriate for the circumstances as when a security incident occurs or as part of the ongoing maintenance of this plan.

As the incident progresses and has more impact (i.e. severity level increases), the escalation process will be used to engage appropriate resources. Incidents should be handled at the lowest escalation level that is capable of responding to the incident with as few resources as possible in order to reduce the total impact, and to keep tight control. The Table below defines the escalation levels with the associated team involvement.

8.7 Migration and Redeployment Plan
#15..
- 837 -
SOLICITATION # CH16012
99 8.7.1 Offeror must describe how it manages the end of life activities of closing down a service to a Purchasing Entity and safely deprovisioning it before the Offeror is no longer contractually obligated to maintain the service, include planned and unplanned activities. An Offeror’s response should include detail on how an Offeror maintains security of the data during this phase of an SLA, if the Offeror provides for redundancy during migration, and how portable the data is during migration.

The end of life activities for each Purchasing Entity will vary based on the type of service they are currently using and the service that they will be moving too. All Purchasing Entities will be notified in advance of any end of life service so that the Purchasing Entity can work with Carahsoft and the service providers to develop a plan exporting and transitioning data.

8.7.2 Offeror must describe how it intends to provide an orderly return of data back to the Purchasing Entity, include any description in your SLA that describes the return of data to a customer.

At all times during the course of the Service, the Purchasing Entities will have access and ownership of their data. The Purchasing Entity is free to download or access the data through a variety of means based on the service at any time during their service period.

8.8 Service or Data Recovery

8.8.1 Describe how you would respond to the following situations; include any contingency plan or policy. a. Extended downtime.

CA APM If unable to resolve in a timely manner, all customers will be notified via email to the registered admin
MAA Customers are kept abreast with progress during incidents, including outages. CA utilizes best-of-breed notification system which enables customer contacts to selfsubscribe to different types of notifications that they would be interested in. A DR plan has been created should the extended down time result in DR declaration.

CA Agile We would failover to our warm data center in order to restore access to the application as quickly as possible. This is done according to our Disaster Recovery Plan ASM If unable to resolve in a timely manner, all customers will be notified via email to the registered admin. Failover from the primary to DR site may be utilized if the extended amount of time warrants declaration of DR.

Google Due the redundant nature of the Google infrastruture there is no expectation of any extended downtime. If a data center in use by any of your end users suffers a catatrophic failure, Google's dynamic health monitoring would just reroute the session to a different data center.

AODocs All the AODocs data is hosted on the Google Cloud Platform infrastructure. AODocs does not manage any physical infrastructure. Due the redundant nature of the Google infrastruture there is no expectation of any extended downtime. If a data center in use by any of your end users suffers a catatrophic failure, Google's dynamic health monitoring would just reroute the session to a different data center.

Virtru Notify users that Virtru service is unavailable, and recommend that they use their offline encryption tools and backups to perform emergency functions.

Salesforce Salesforce has maintained high levels of availability across all Salesforce instances since inception. As the only on-demand vendor to provide daily service-quality data on a public Web site (http://trust.salesforce.com), Salesforce proves that we are the leader in availability. And by making its track record completely transparent, Salesforce proves we are worthy of our customers’ trust. To ensure maximum uptime and continuous availability, Salesforce provides the best #15..
- 838 -
SOLICITATION # CH16012
100
redundant data protection and most advanced facilities protection available, along with a complete data recovery plan—all without affecting performance.

Salesforce uses commercially reasonable efforts to make its on-demand services available to its customers 24/7, except for planned downtime, for which Salesforce gives customers prior notice, and force majeure events. Excellent availability statistics are critical to Salesforce's customers’ success and to the success of Salesforce as a company. Live and historical statistics on the Salesforce system performance are publicly published at http://trust.salesforce.com/trust/status.

The persistence layer underlying Salesforce Platform is proven database technology that powers all of Salesforce’s products today, serving more than 150,000 organizations and over 4 billion transactions per day with an average request response time of less than .25 seconds, all with an average up time of 99.9+ percent.

ServiceNow ServiceNow’s data centers and cloud-based infrastructure have been designed to be highly available. All servers and network devices have redundant components and multiple network paths to avoid single points of failure.

ServiceNow leverages this Advanced High Availability (AHA) architecture for customer production instances in several ways:
•In the event of the failure of one or more infrastructure components, service is restored by transferring the operation of customer instances associated with the failed components to the peer data center.
•Before executing required maintenance, ServiceNow can proactively transfer operation of customer instances impacted by the maintenance to the peer data center. The maintenance can then proceed without impacting service availability.

This approach means that the transfer between active and standby data centers is being regularly executed as part of standard operating procedures – ensuring that when it is needed to address a failure, the transfer will be successful and service disruption minimized.

RTO is 2 hours. RPO is 1 hour.

See the “ServiceNow Security, Operations, and Compliance White Paper” included with this response for more information.

QTS In the rare instance of failure of the commercially-delivered electrical power, back-up systems at each QTS data center start instantaneously. Batteries provide power for the first 12-15 seconds while on-site diesel generators spin up to synchronize their output with the battery-supplied feed.

Once on-line, the generators can supply power indefinitely, and the IT equipment on the data floor as well as critical infrastructure components are all unaffected.

While the basics of reliable power seem simple, none of this is taken for granted. Battery banks are sized to provide 15 minutes of power, well beyond the 15 seconds that it typically takes generators to come on line. Inspections are conducted daily and quarterly to ensure that the battery banks are prepared at all times. The diesel generators are inspected by running them each month to circulate and condition the fuel, then running them annually under the full simulated load of the current IT equipment on the data center floor. Factory technicians inspect and maintain the generators semi-annually. Quantities of fuel sufficient to support 48-72 hours of #15..
- 839 -
SOLICITATION # CH16012
101
operation without resupply are stored on-site, and re-supply contracts are in place with multiple providers to have fuel delivered as soon as the generators come on line.

SAP Ariba Ariba has implemented two sites within each region. In North America the sites are currently located in San Jose, California and Sterling, VA. In Europe the sites are located in St. Leon-Rot, Germany and Amsterdam, Netherlands.

The act of failing over from one site to another has a design goal for its Recovery Time Objective (RTO) of no more than four hours. The design goal for the Recovery Point Objective (RPO) is five minutes.

Disaster recovery options are included for all Ariba Cloud Services. In the event of a fail-over to the disaster recovery site, no customer changes are required as all URLs that customers use to reach the applications will continue to work. Ariba will notify customers via their email addresses in the event of unplanned downtime.

Internally, Ariba uses a documented system recovery plan that outlines the approach and steps for recovering the applications. This document defines roles and responsibilities in the event of disaster:

Local Ariba staff maintains the hardware remotely.

Ariba maintains the application software.

Processes are in place to keep database and fileservers in sync between primary and backup data centers.

The failover process of all parts of the infrastructure is automated.

In the event of a catastrophe, Ariba will declare the primary data center “down” and locally the script will be run to switchover and start the applications at the remote data center.

Ariba tests power outage backup scenarios and the Disaster Recovery Plan on a periodic basis to ensure it is up-to-date, successful, and effective.

Fieldglass Customers are notified of unscheduled downtime immediately via email.

Updates are frequently emailed out until the issue has been resolved.

Customers are notified of scheduled downtime 5 days prior via email.

Hanna In case of a disaster depending on the infra & technical architecture defined during HEC technical assessment workshop, systems could fail over to HA node. If DR (optional offering) is selected, the systems could move to secondary DC that where contracted with a DR option & HEC declaring DR.

Regular system downtime for patching etc. is planned in advance with the customer and typically follows customers maintenance schedules.

Hybris SAP Hybris offers continuous 24/7 systems monitoring in-place, which automatically notifies you, SAP Hybris support and optionally your designated implementation partner in the event of any monitored system problem. Tools used for monitoring include HP Sitescope and Nagios. This service ensures quick response times to emergency problems. Standard monitors are in place for basic site availability. Customer can utilize additional 3rd party monitoring services should it wish to add additional monitoring. Upon request, monthly or quarterly reviews are offered to review performance over the period. When a system problem occurs, in addition to monitor alerts sent to the customer, SAP Hybris support may send emails to a designated customer notification email address to provide status, updates or further detailed information. In addition, a support ticket would be created which would include such information.

SuccessFactors Our infrastructure architecture is designed with high availability in mind, and engineered for resiliency. All major components are redundant, including power, HVAC, fire suppression, and the physical components of our network.

Production data centers have strict access controls, and are continuously #15..
- 840 -
SOLICITATION # CH16012
102
staffed and monitored to help prevent acts of sabotage or vandalism. All production data centers are ANSI/EIA/TIA-942Tier III/IV facilities, and are ISO 27001 certified.

Production data centers are also geographically dispersed to help prevent a single event from affecting more than one data center. In the event a production data center has an outage we failover to an alternate data center in the same geographic region to minimize impact to customers.

Our Cloud Operations teams are also geographically dispersed, working in offices in the US, Europe, South America, and India. Should an office be impacted by an environmental event or pandemic, other offices can continue operations.

VMware VMware vCloud Air leverages proven VMware High Availability technology to minimize the risk of extended downtime. In the unlikely event that an extended outage does occur, VMware will support guide the customer's efforts to restore services to an alternative vCloud Air or vCGS datacenter. We will also provide remedies in the form of service credits as outlined in the following VMware vCloud Air, vCGS, Vmware AirWatch and VMware Identity Manager SLAs.

FireEye FireEye has a mature and documented Business Continuity Plan and Disaster Recovery Plan that follows industry best practices that is reviewed and tested annually. The BCP/DRP plan has RPO/RTO timing criteria and other critical business requirements that is available for review if requested in the future.

VirtueStream In the event of extended downtime affecting production systems, Virtustream and customer would jointly evaluate the situation and mutually decide on failover declaration, whereby Virtustream would then execute the failover plan to restore customer operations to the secondary data center. b. Suffers an unrecoverable loss of data.

CA APM Customer data is isolated within dedicated schemas with nightly backups. The maximum data loss would be the previous 24hrs.

MAA MAA data is backed up fully daily. The maximum data loss would be the previous 24hrs.

CA Agile We have implemented a physical standby database in both hot/Live site as well as warm/standby sites using Oracle Data Guard with realtime apply to achieve the stated recovery objectives.

ASM Customer data is isolated within dedicated schemas with nightly backups. The maximum data loss would be the previous 24hrs.

Google Google performs real time data replication to avoid unrecoverable data loss. Customers that configure the retention policies in Google Apps Vault can ensure that end users can not inadvertently or maliciously delete email in advance of any defined retention schedules.

AODocs AODocs data are stored in Google AppEngine and Google Drive:
- The data stored in Google AppEngine are backed up every day.

The application is hosted on Google Cloud Platform infrastructure and benefits from the network security described in this whitepaper: https://cloud.google.com/files/Google-CommonSecurityWhitePaper-v1.4.pdf - Data stored in Google Drive is managed by Google within its multiple servers location. Altirnao does not manage this infrastructure. Google Drive can be backed up by third party tool such as Backupify.
#15..
- 841 -
SOLICITATION # CH16012
103
Virtru Notify users that data within Virtru is unavailable, and recommend that they use their offline encryption tools and backups to perform emergency functions.

Salesforce To maximize availability, the service is delivered using multiple world-class data centers supporting primary and replicated disaster recovery instance, plus a separate production-class lab.

The infrastructure utilizes carrier-class components designed to support millions of users.

Extensive use of high-availability servers and network technologies, and a carrier-neutral network strategy, help to minimize the risk of single points of failure, and provide a highly resilient environment with maximum uptime and performance.

The Salesforce services are configured to be N+1 redundant at a minimum, where N is the number of components of a given type needed for the service to operate, and +1 is the redundancy. In many cases, Salesforce has more than one piece of redundant equipment for a given function.

ServiceNow Please see our response for 8.8.1a
QTS QTS cloud developed a OPS_POL_54_QTS-FC_Contingency_Planning_Policy (v1.2 – 2/6/14), detailing the policies, procedures, roles and responsibilities of key personnel in the event of an emergency and/or disaster.

SAP Ariba Our SLA provides tenant remuneration for losses they may incur due to outages within the infrastructure.

Fieldglass The maximum data loss timeframe is guaranteed to be less than three hours; however, offsite backups are performed every 15 minutes so the timeframe is actually much shorter.

Hanna Data loss is defined by RPO & RTO. HEC has a maximum RPO of 30 mins & RTO of 12 hours.

Hybris For a cloud solution, data backup and restore is handled by SAP. SAP shall follow its archiving procedures for Customer Data as set out in the SAP Hybris Commerce, cloud edition Services Description. In the event of any loss or damage to Customer Data, SAP shall use commercially reasonable efforts to restore the lost or damaged Customer Data from the latest back- up of such Customer Data maintained by SAP in accordance with the archiving procedure described in its SAP Hybris Commerce, cloud edition Services Description. SAP shall not be responsible for any loss, destruction, alteration or disclosure of Customer Data caused by any third party, (except those third parties sub- contracted by SAP to perform services related to Customer Data maintenance and back-up).

SuccessFactors Our SLA provides tenant remuneration for losses they may incur due to outages within the infrastructure.

VMware VMWare IaaS Services
As discussed in detail in section 8.8.2.1.a below, VMware IaaS services are architected to minimize the risk of unrecoverable data loss. In addition option services such as vCloud Air Data Protection can be leveraged to further mitigate any risk. VMware’s vCloud Air policy regarding data loss can be found in the vCloud Air Terms of use. VMware’s vCloud Government Service policy regarding data loss can be found in the vCloud Government Service Terms of Service AirWatch Hosted Services
AirWatch Business Continuity and Disaster Recovery strategies include data and hardware redundancy, network configuration redundancy and backups, and robust, regular testing exercises. AirWatch features active-passive configurations for high availability and redundancy with all components made to failover with minimal downtime. Load balancing capabilities are deployed across multiple data centers to ensure timely server recovery. AirWatch also incorporates replication technology featuring SQL log shipping or network SAN byte replication to prevent data loss.
#15..
- 842 -
SOLICITATION # CH16012
104
• Due to FOIA restrictions and the competitive nature of the EMM marketplace, we cannot provide specific details regarding recovery strategies or timelines.

VirtueStream Multiple copies of customer data exist in multiple datacenters – in the primary datacenter, there is the production copy and a local backup. The local backup is replicated to a secondary datacenter and the production datastores associated with coreVM’s are also replicated at regular intervals to the secondary datacenter. If NASPO feels that additional steps/redundancy are required to maintain recoverability, Virtustream is open to discussion. c. Offeror experiences a system failure.

CA APM Disaster to the CA Technologies corporate network in New York will not affect customers' service. Secondary services, such as domain name services will be routed through the secondary CA Technologies network in Illinois. CA has a BCP plan in place to direct its services. The SaaS environment is separate from the CA corporate network and a service specific disaster recovery plan is in place.

MAA Customers would be notified in advance, and given ample time to retrieve their data.

CA Agile We enable all of our employees with the ability to be able to work remotely and provide remote network access to ensure business functions can continue. All corporate infrastructure has redundate systems that can be utilized in the event of failure.

ASM Customers would be notified in advance, and given ample time to retrieve their data manually or via the ASM API.

Google Google's corporate network is separate from it's production infrastructure. Google is a global organization that has support personnel located at key office installations and tests our business continuity programs annually. These tests are to validate that if corporate headquarters is off the grid that criticial business functions can be picked up by other staff.

AODocs AODocs runs 100% over Google's Cloud Platform infrastructure that is naturally redundant, and all AODocs internal systems such as technical support, email, collaboration, source code hosting, etc are run on cloud based services. AODocs personnel can work from any physical location without any business disruption and they do work from remote locations on a regular basis. If one of the AODocs facilities were to be temporarily unavailable due to a natural disaster, AODocs personnel would be able to perform their work without significant interruption.

Virtru Notify users that Virtru is or will be permanently unavailable, and recommend that they use their offline encryption tools and backups to perform emergency functions.

Salesforce Please see response to items a. and b. above and d. and e. below.

Salesforce has documented Disaster Recovery and Business Continuity plans for critical business functions. The Disaster Recovery and Business Continuity plans are tested at least annually. A post mortem documenting the results of the disaster recovery tests can be provided to customers with a signed NDA in place.

Business continuity plans are updated each year, including the list of business processes, recovery time objectives, and key resources. Senior management is included in this process.

Business continuity plans are exercised on an annual basis. Action items and lessons learned are tracked from each incident and exercise conducted. Action items are prioritized and tracked until #15..
- 843 -
SOLICITATION # CH16012
105
closed. Salesforce has developed additional procedures, processes and plans, including a Pandemic plan.

Salesforce also recommends customers also devise their own backup strategy for their data, as there is a fee for customers to request for restoration from Salesforce backups. Salesforce provides multiple ways for our customers to obtain periodic backups of its data. We offer a weekly export service (WES) for those customers requiring a local backup copy of their data or a data set for import into other applications (such as an ERP system). Salesforce also supports data replication, which allows customers to store and maintain a local, separate copy of their organization's Salesforce data including the META data (logins, etc.) for specialized uses, such as data warehousing, data mining, custom reporting, analytics, and integration with other applications. Data replication provides customers with local control and the ability to run large or ad hoc analytical queries across the entire data set.

ServiceNow Please see our response for 8.8.1a
QTS QTS has established an alternate processing site including necessary agreements to permit the resumption of information system operations for essential missions and business functions within [Assignment: organization-defined time period consistent with recovery time objectives] when the primary processing capabilities are unavailable.

The QTS cloud hosting facilities in Richmond, VA (RIC1) and Atlanta, GA (ATL1) have agreements with each other to restore services in the event that one of the sites becomes unavailable. Both facilities are QTS managed facilities and are always available to meet the recovery time and recovery point objectives for the system. Agreements are in place to allow QTS to access resources at the alternate processing site physically and via the QTS client network.

Richmond and Atlanta Datacenters are independently operated datacenters that offers disaster recovery services to customers. The contingency approach ensures that hardware and software components needed for recovery efforts are already at the alternate facility.

QTS ensures that equipment and supplies required to resume operations are available at the alternate site or contracts are in place to support delivery to the site in time to support the organization-defined time period for resumption.

Network circuits, networking infrastructure, storage, virtual servers, and management components required to resume operations are available and configured at the alternate site.

SAP Ariba "Ariba modules automatically persist object state to the underlying relational database as users work. When a user performs a significant action, for example, adding an item to an order, performing an approval, etc., the Ariba application tier is aware of the change and automatically saves the relevant object state to the database. This provides transparent persistence of the user's work without the user having to “save” or “submit” changes to avoid losing work in the event of failure or session timeout. This in turn means that the loss of any connections between the tiers or failure of one of the tiers will not leave data in an inconsistent state.

While object data is cached in the middle tier for performance, the official record is stored in the database to avoid any potential data loss or corruption.

This means that recovery for Ariba modules is implicit and automatic. When the modules start, they retrieve the state of the various business objects from the underlying database as needed for user sessions. If an Ariba module fails, there is no special application recovery process other than restarting the module. The Ariba module will automatically connect to the database and as clients login and commence working, all required business object data will be retrieved from the database."
#15..
- 844 -
SOLICITATION # CH16012
106
Fieldglass Fieldglass’ business continuity program (BC Program) was established to ensure that, after a complete disruption of its Chicago and/or Naperville, Illinois offices, all employees and the functions for which they are responsible, are capable of continued operations in less than one business day. This includes office space as well as all equipment and supplies needed to continue its operations without an interruption in customer services.

Fieldglass’ BC Program is integrated into all facets of the company due to the criticality of the product that it sells and supports. As such, representatives from all areas of the company are responsible for ensuring that they are familiar with their duties and responsibilities during a disruption and how to recover their areas after the disruption occurs.

Hanna Service Level Credits
Where SAP fails to meet a Service Level, SAP will be liable to Customer for the corresponding Service Level Credit as set out in this section. The Service Level Credit is calculated as the sum of the Service Level Credits for both DEV/QAS and PRD for the TA Service Level defined in section 5.1 above.

SAP will deduct the amount of any Service Level Credits owed to Customer from the next invoice (or, if there is no such invoice, by bank transfer to such bank account as Customer may specify in writing).

Customer agrees that under no circumstances will the total maximum Service Level Credits: (i) for any one month, exceed 100% of the Service Fee for that month; and, (ii) for any given contract year, exceed in the aggregate an amount equal to one- third of the annual Service Fee charged for the contract year (or one third of the total Service Fee charged if the Term as defined in the applicable Order Form is less than one (1) year). Customer acknowledges that the Service Level Credits defined hereunder are the sole remedy for SAP’s failure to meet the specified Service Level.

5.6 Termination for Service Level Failure
Customer may terminate the applicable Order Form with thirty (30) day’s termination notice in writing to SAP, if SAP misses a Service Level as specified in this Supplement for three (3) months in sequence. Customer may exercise this termination right only within thirty (30) days after receipt of the respective Service Level report that documents the applicable Service Level failure that would cause such termination right to accrue in Customer.

Hybris SAP isn't likely to go out of business in the near future. However, our contract states the following: Upon expiration or termination of the Agreement, SAP may destroy or otherwise dispose of any of Customer Data in its possession unless SAP (i) is requested by Customer to extend the term of the Order Form as permitted in the Order Form to allow Customer to retrieve Customer Data, or (ii) receives, no later than thirty (30) days after the effective date of the termination of this Agreement, a written request for the delivery to Customer of the then most recent back-up of the Customer Data. SAP shall use reasonable commercial efforts to deliver the back-up to Customer within thirty (30) days of its receipt of such a written request, provided that Customer has, at that time, paid all fees and charges outstanding and owed at termination. Customer shall pay all reasonable fees and expenses incurred by SAP in returning or disposing of the Customer Data.
#15..
- 845 -
SOLICITATION # CH16012
107
VMware AirWatch Hosted Services
AirWatch Business Continuity (BC) strategies include clear policies and procedures as well as robust, regular testing exercises. Our BC recovery strategies for support and maintenance include provisions for workforce mobility, secondary support locations, and defined leadership roles and responsibilities for all recovery staff.
• Multiple Support Locations – AirWatch maintains a Global Support Team, which spans multiple offices around the world. This ensures that if a disaster occurs at one of our offices, our remaining Support Team members can provide assistance to our customers. • Mobile Workforce – Should an AirWatch office become inaccessible, all AirWatch employees are equipped with laptops and can access necessary internal resources. • Defined Leadership – During a disaster, roles and responsibilities are assigned to key personnel, and a recovery manager provides leadership and documents the continuity process. Additionally, emergency contact numbers for key personnel are provided for all vendors and support staff.

VirtueStream Virtustream has built redundancy into all layers of the physical infrastructure in order to deliver SLAs up to 99.999% and mitigate the risks of a system failure. Additional levels of protection include regular system backups (stored locally and replicated offsite) as well as automatic storage replication to a secondary data center to facilitate the ability to restore operations in the event of a system failure.
d. Ability to recover and restore data within 4 business hours in the event of a severe system outage.

CA APM CA provides an SLA of 99.8% uptime, which can result in unforeseen outages of ~1.5 hours per month. In the event of a failure to meet a SLA threshold , Customer is entitled to a number of days of credit.

MAA Using DB clustering technologies, multiple copies of data are maintained helping in recovery of the data.

CA Agile We can meet this requirement.

ASM Data is replicated from the primary (production) to secondary (DR) continuously. In the event of catastrophic loss of live data a failover to the DR site would be necessary.

Google Google's RTO is zero.

Virtru See 'Virtru Incident & Breach Policy'
Salesforce Customer data, up to the last committed transaction, is replicated to disk in near-real time at the designated disaster recovery data center, and backed up at the primary data center. Backups are performed daily at primary data center facility without stopping access to the application.

For business continuity purposes, Salesforce supports disaster recovery with a dedicated team and a 4 hour recovery point objective (RPO) and 12 hour recovery time objective (RTO).

Additional details can be provided with the execution of an NDA between Salesforce and your Agency.

ServiceNow Please see our response for 8.8.1a
QTS As defined in QTS OPS_POL_54_QTS-FC_Contingency_Planning_Policy (v1.2 – 2/6/14), QTS cloud has identified the following potential accessibility problems:
• Network Outage – Short Term
• Network Outage – Long Term
• Facility Power or Environmental Outage – Short Term
#15..
- 846 -
SOLICITATION # CH16012
108
• Facility Power or Environmental Outage – Long Term
• Serious Facility Damage – Short Term
• Serious Facility Damage – Long Term or Permanent
Both the RIC2 and ATL1 QTS cloud Infrastructure is an Active/Active system and can operate independently of each other indefinitely.

If QTS cloud determines that any single event (network, power, facility damage) presents a longterm threat, QTS immediately conducts an emergency meeting and plan to remedy or relocate infrastructure to a viable facility."
SAP Ariba Ariba has implemented two sites within each region. In North America the sites are currently located in San Jose, California and Sterling, VA. In Europe the sites are located in St. Leon-Rot, Germany and Amsterdam, Netherlands.

Fieldglass Fieldglass offers the following:
• RTO = No more than 48 hours
• RPO = No more than 6 hours.

Warm site replica with 100% processing power of original site. Disaster recovery site can run without need from resumption.

Hanna See our RTO and RPO info below
Hybris For a cloud solution, data backup and restore is handled by SAP.SAP shall follow its archiving procedures for Customer Data as set out in the SAP Hybris Commerce, cloud edition Services Description. In the event of any loss or damage to Customer Data, SAP shall use commercially reasonable efforts to restore the lost or damaged Customer Data from the latest back- up of such Customer Data maintained by SAP in accordance with the archiving procedure described in its SAP Hybris Commerce, cloud edition Services Description. SAP shall not be responsible for any loss, destruction, alteration or disclosure of Customer Data caused by any third party, (except those third parties sub- contracted by SAP to perform services related to Customer Data maintenance and back-up).

Both IDS and IPS are included the SAP Hybris Commerce, Cloud Edition The security infrastructure includes firewall security and hardened security policies on all servers. Log management procedures are in-place for log review for firewall, applications, network devices, including file-integrity management. SAP Hybris utilizes technologies from leading security firms for Log Management and File Integrity Management. SAP Hybris employs twofactor authentication across its network. SAP Hybris undergoes vulnerability and penetration testing. SAP Hybris validates against requirements for PCI DSS 2.0.

The infrastructure also includes Web Application Firewalls and DDoS Mitigation Services.

In addition, security policies and change management policies are in-place ensuring that all access and changes to customer systems and information is accessible only by SAP Hybris staff with access authorization.

Security of the software application which is controlled by the Customer (or its implementation partner) remains the responsibility of the Customer.

Upon expiration or termination of the Agreement, SAP may destroy or otherwise dispose of any of Customer Data in its possession unless SAP (i) is requested by Customer to extend the term of the Order Form as permitted in #15..
- 847 -
SOLICITATION # CH16012
109
the Order Form to allow Customer to retrieve Customer Data, or (ii) receives, no later than thirty (30) days after the effective date of the termination of this Agreement, a written request for the delivery to Customer of the then most recent back-up of the Customer Data. SAP shall use reasonable commercial efforts to deliver the back-up to Customer within thirty (30) days of its receipt of such a written request, provided that Customer has, at that time, paid all fees and charges outstanding and owed at termination. Customer shall pay all reasonable fees and expenses incurred by SAP in returning or disposing of the Customer Data.

SuccessFactors RPO and RTO are defined in the general terms and conditions VMware VMware Response:

VMware IaaS Services
VMware DR service.

We can discuss specific recovery information under NDA to participating entities during task order negotiations as required.

AirWatch Hosted Services
We can discuss specific recovery information under NDA to participating entities during task order negotiations as required.

VirtueStream Basic Plus microVM’s are designed for mission-critical workloads and for systems that require DR, addition of reserved compute capacity in a secondary datacenter (i.e. Reserve uVM) and automatic storage replication to facilitate the ability to restore operations in the event of an outage at the primary data center. RPO and RTO for core microVM's are 15 minutes and 2 hours respectively.
e. Describe your Recovery Point Objective (RPO) and Recovery Time Objective (RTO).

CA APM Recovery Point Objective (RPO): Maximum data loss: 24 hours Data that is uploaded, but not backed up within the 24 hours may have to be reentered Recovery Time Objective (RTO): 72 hours
MAA Recovery Point Objective (RPO): Maximum data loss: 24 hours Data that is uploaded, but not backed up within the 24 hours may have to be reentered Recovery Time Objective (RTO): 72 hours
CA Agile RTO: 2 hours
RPO: 12 hours
ASM Recovery Point Objective (RPO): Maximum data loss: 24 hours Data that is uploaded, but not backed up within the 24 hours may have to be reentered Recovery Time Objective (RTO): 24 hours
Google RPO/RTO objectives are zero hours.

AODocs AODocs is hosted on Google App Engine which is designed to be highly available. Our standard SLA is 99.5% . Specific financial penalties can be discussed as part of the AODocs license contract.

Google App Engine is running in multiple datacenters on multiple continents. #15..
- 848 -
SOLICITATION # CH16012
110
AODocs data is backed up on a daily basis, and backups are stored on the highly redundant Google Cloud Storage. Backups are retained for at least 90 days Virtru See 'Virtru Incident & Breach Policy'
Salesforce For business continuity purposes, Salesforce supports disaster recovery with a dedicated team and a 4 hour recovery point objective (RPO) and 12 hour recovery time objective (RTO).

Additional details can be provided with the execution of an NDA between Salesforce and your Agency.

ServiceNow Please see our response for 8.8.1a
QTS QTS DRaaS will support all of your mission-critical application needs and enables you to achieve a RTO within minutes and RPO of just seconds.

SAP Ariba See above
Fieldglass Fieldglass offers the following:
• RTO = No more than 48 hours
• RPO = No more than 6 hours.

Warm site replica with 100% processing power of original site. Disaster recovery site can run without need from resumption.

Hanna For disaster Recovery, SAP HEC offers the following RTO & RPO

1. HANA Database
RTO - 12 hours, RPO - 30 minutes

2. Sybase ASE Database
RTO - 12 hours, RPO - 30 minutes
Hybris In the case of catastrophic failure to your primary production data center, we offer anywhere from best efforts to enhanced DR solutions. Enhanced DR options translates to either RTO of 8 hours and RPO of 1 hour for a warm standby solution, or RTO of 30 minutes to RPO of 30 minutes for a hot-site.

Customers have 3 choices in DR sites:
• as part of our base offering, best effort DR, we would bring your site back up in an alternate datacenter which is a minimum of 500 miles away from the primary datacenter. We would actually push your project server images to the alternate datacenter, but we would still need to set up your various interfaces, 3rd party services, etc, which could day a few days.
• 2nd would be to have your DR site managed by hybris Managed Services. • 3rd would be to use your own DC as a DR site, We would need to set up a technical call with our combined teams to synch up on the replication strategies.
• Our DR services utilizes VMware Site Recovery Manager with storage-based replication.
• For enhanced DR, we would include one annual DR test, but more can be added if required. It also includes a customer specific DR and Business Continuity plan.

SuccessFactors Within the contract/subscription agreement, there are provisions for data recovery in the event of disaster. SLA's for RTO (Recovery Time Objective) and RPO (Recovery Point Objective) are defined in the contract. We will prioritize data recovery based upon our contractual obligations for each customer, concerning RTO and RPO.
#15..
- 849 -
SOLICITATION # CH16012
111
VMware VMware IaaS Services
Provide the RPO and RTO for DR, point to SLAs for others.

RTO – 4 hours or less
RPO – 15 minutes to 24 hours – time is configurable by user but there is a bandwidth consideration for shorter replication times.

VMWare AirWatch
AirWatch Business Continuity (BC) and Disaster Recovery (DR) strategies include data and hardware redundancy, network configuration redundancy and backups, and robust, regular testing exercises.
• Business Continuity – Our BC recovery strategies for support and maintenance include provisions for workforce mobility, secondary support locations, and defined leadership roles and responsibilities for all recovery staff.
o Multiple Support Locations – AirWatch maintains a Global Support Team, which spans multiple offices around the world. This ensures that if a disaster occurs at one of our offices, our remaining Support Team members can provide assistance to our customers. o Mobile Workforce – Should an AirWatch office become inaccessible, all AirWatch employees are equipped with laptops and can access necessary internal resources. o Defined Leadership – During a disaster, roles and responsibilities are assigned to key personnel, and a recovery manager provides leadership and documents the continuity process. Additionally, emergency contact numbers for key personnel are provided for all vendors and support staff. • High Availability and Disaster Recovery – AirWatch features active-passive configurations for high availability and redundancy with all components made to failover with minimal downtime.

Load balancing capabilities are deployed across multiple data centers to ensure immediate server pick up, ensuring zero end user downtime. AirWatch also incorporates replication technology featuring SQL log shipping or network SAN byte replication to prevent data loss.

Due to FOIA restrictions and the competitive nature of the EMM marketplace, we cannot provide specific details regarding recovery strategies or timelines.

VirtueStream Virtustream’s solution will have a Recovery Point Objective (RPO) of 2 Hours and Recovery Time Objective (RTO) of 30 Minutes for all disaster recovery enabled workloads.

The Virtustream Cloud has been designed to deliver continuous operations. The platform is architected to be highly available, with SLAs as up to 99.99%. All infrastructure systems are run in at least an N+1 model and have been designed with no single points of failure. All of the data center facilities and infrastructure has redundant power supplies connected to separate circuits from separate power feeds. Every device is connected via redundant pathways at a LAN/SAN/WAN layer.

State of Utah can choose any of our FedRAMP Data Centers to host their applications and the DR location. For illustrative purposes, we have assumed the primary data center will be USDC02 and DR data center would be USDC01.

The State of Utah solution has been designed with all workloads running in our San Francisco (USDC02) data center. In the unlikely event of a disaster at the USDC02 facility, State of Utah’s production services will be brought back online in the secondary data center located in Northern Virginia (USDC01). The target RPO is 30 minutes for virtual machines with a 2 hour RTO. All production workloads in the primary node will have reserved compute capacity in the secondary node when deployed using Reserve µVMs and replicated storage. All virtual machine disaster recovery services are based on storage replication. The primary and secondary data centers are interconnected by diverse private 10 Gbps Sonet rings from separate Tier 1 providers – AboveNet and Level 3. Data is replicated asynchronously and continuously between the two data centers. #15..
- 850 -
SOLICITATION # CH16012
112
Virtustream also performs backups on a nightly basis and replicated offsite for further protection.

The retention policies are determined by State of Utah per landscape and will be documented during the on-boarding process.

Figure 2. Integrated Disaster Recovery

8.8.2 Describe your methodologies for the following backup and restore services:
a. Method of data backups
CA APM Backups are stored on disk only
MAA Backups are stored on disk only, sync'd between data centers for DR purposes.

CA Agile Full database backups are taken once per week, both to onsite and offsite (our warm data center, not a separate 3rd party) systems. Incremental backups are run nightly.

Backups are retained for 21 days after which time they are simply aged out and overwritten.

In addition, snapshots of database transactions are taken every hour to an disaster recovery site allowing for emergency disaster recovery with maximum of 1.5 hours of data loss due to catastrophic onsite failure (fire in cage, natural disaster, etc. at data center). Backups are tested monthly and a full disaster recovery process to the offsite application cluster is tested semi-annually.

ASM Backups are stored on disk only, sync'd between data centers for DR purposes.

Google Keeping in mind that Google stores customer data in encrypted chunks or shards on several servers at several data centers backups should not be considered customer specific. Daily backups of each server are performed to encrypted tapes at each data center.

Virtru We perform live database replication to a separate redundant database in Oregon and offline as well in a separate Amazon AWS Region
Salesforce Customer data, up to the last committed transaction, is replicated to disk in near-real time at the designated disaster recovery data center, backed up at the primary data center, and then cloned to the disaster recovery data center. Disaster recovery tests verify our projected recovery times and the integrity of the customer data.
#15..
- 851 -
SOLICITATION # CH16012
113
Backups are performed daily at each data center facility without stopping access to the application. Backup cloning is transmitted over an encrypted network (our MPLS network across all data centers). Tapes never leave our secure data center facilities, unless they are to be retired and destroyed through a secure destruction process.

The backup retention policy is 90 days (30 days for sandboxes). Deleted / modified data cannot be recovered after 90 days (30 days for sandboxes). If customers want a longer retention, they can use the weekly export feature available in the system.

ServiceNow While Advanced High Availability as described in 8.8.1 is the primary means to recover data and restore service in the case of a service disruption, in certain cases it is desirable to use ServiceNow’s more traditional data backup and recovery mechanism. This data backup and recovery system works in concert with AHA and acts as a secondary recovery mechanism.

ServiceNow stores production instances in two geographically separate regional data centers, with sub-production instances hosted in a single data center. Backups of the two production databases and the single sub-production database are taken everyday for all instances throughout the private cloud infrastructure.

The backup cycle consists of four weekly full backups and the past 6 days of daily differential backups that provide 28 days of backups. All backups are written to disk, no tapes are used and no backups are sent off site. All the controls that apply to live customer data also apply to backups. If data is encrypted in the live database then it will also be encrypted in the backups.

Regular, automated tests are run to ensure the quality of backups. Any failures are reported for remediation within ServiceNow.

QTS QTS operates and manages high-performance, multi-tenant platforms on which customer system and data file backups can be created and retained for rapid restore of files/volumes. Data can be stored on tape or disk, depending on QTS location, and can reside locally in a QTS data center or off-site. If there is a need for off-site data retention, the backup medium is tape. QTS can also replicate the backed up data from the primary site to a secondary (remote) site to address geographically remote storage and DR requirements.

Secure: QTS utilizes a separate, secure gigabit backup network with private VLAN security.

Physical security is monitored using a priority badge access system to limit access to the tape libraries, disk arrays, servers, and network infrastructure. From the application, the backup software controls access to the data stored on tape, and access security is limited to QTS staff only.

Scalable: QTS backup systems are highly scalable, and capacity can be quickly added as needed.

SAP Ariba Our primary data center is Equinix, in San Jose; CA. Ariba's backup data center is CenturyLink in Sterling, VA. We make many copies of our data to insure no data loss happens:

Data is stored in databases on high-availability storage disk-based system in primary datacenter
Data is copied as backup to high-availability disk storage in the primary datacenter
Data is copied to tape and kept in the primary data center Data is copied to tape and kept in an off-site tape storage facility (Iron Mountain) in Union City
Data is replicated to databases high-availability storage disk-based system in the backup datacenter
Data is copied as backup to high-availability disk storage in the backup datacenter
Data is copied to tape and kept in the backup datacenter #15..
- 852 -
SOLICITATION # CH16012
114
Data replication to the remote datacenter happens in near-real time.

Data is backed up to the backup disks multiple times a day (in each location) Data is written to tape once a day (in each location)
These processes and procedures are audited against the ISAE 3402 standard. Our auditor verifies we can restore from these backup methods.

Fieldglass All data in the Fieldglass system is stored within our clustered databases. Full backups daily, 15 minute incremental backups, and replication from center to center of backup data continuously.

Hanna The backup of SAP systems in HEC is executed on EMC DataDomain deduplication storage. This is done using EMC Networker for backup and administration. The backups on Site A are also replicated to remote Site B.

SuccessFactors Data backup is undertaken as follows: storage system snapshots every 24 hours, nightly incremental database backups, and weekly full database backups. These are all standard and provided as part of our hosted service.

All backups are stored on disk in the primary and backup data center facilities for 30 days.

Database backups are encrypted and stored at a customer’s “primary” location, as well as the “alternate” warm-site location for redundancy and disaster recovery purposes. If the customer’s primary location is defined as the data center in Amsterdam, NL, then copies of their database would be encrypted and streamed in a secure manner to the data center based in St Leon-Rot, DE, and vice versa. The backup is stored on Storage Area Network systems encrypted with AES-256 bit by encryption appliance. Backups are retained for 30 days. Our Log File Retention Policy is designed to meet or exceed the most stringent industry standards. Archiving of data is the customer's responsibility.

VMware VMware will provide the following services to participating entities with respect to Data Recovery:
•Data protection, such as routine backups, for the vCloud Air infrastructure, including top-layer management and user-management interfaces owned and operated by VMware. •Data and infrastructure restoration for the vCloud Air infrastructure, including top-layer management and user-management interfaces owned and operated by VMware.

Participating entities will be responsible for the following services with respect to Data Recovery:
•Data protection, such as routine backups, for the data and content accessed or stored on vCloud Air virtual machine’s or storage devices, configuration settings, etc. •Data, content, virtual machine and configuration restorations for assets accessed or stored on your vCloud Air account.

Data Protection is an optional service that provides secure, image-based backup and recovery capabilities that enable you to protect important virtual machine data and content hosted in your vCloud Air IaaS environment. Through the Data Protection administration interface available in the vCloud Air Console, vApps and their virtual machine members can be selected for policy-based backup and recovery operations.

Data Protection feature subscription and activation may be requested via My VMware and is subject to additional service fees based on the amount of backup data capacity. Backup data capacity for the service is measured in front end terabytes (FETB). Once activated, vApps and their virtual machine members may be registered and unregistered with Data Protection features on a self-service basis through the vCloud Air Console.

As part of this service, VMware will:
•Implement and maintain central service components (backup software appliances, backup and archival storage media and associated network topologies) needed to support Data Protection features.
•Perform routine configuration, maintenance and optimization services on behalf of the Data Protection environment and in conformance with industry best practices. #15..
- 853 -
SOLICITATION # CH16012
115
•Allocate requisite backup storage based on capacity selections made at the time of subscription enrollment.
•Guarantee storage locality per geographical region for all backup data. •Provide necessary Data Protection service reporting as it is requested.

Participating entities will be responsible for:
•Subscribing to Data Protection as an add-on feature via My VMware and selecting an amount of backup storage capacity commensurate with your requirements. •Creating custom backup protection policies that may include, but are not limited to: affinity settings per VDC, scheduling, and retention periods.
•Registering and unregistering individual vApps and their virtual machine members for scheduled backups using Data Protection.
•Performing any on-demand backups per vApp and its virtual machine members. •Performing in-place or out-of-place restores per vApp and/or individual virtual machine. •Managing any in-guest recovery tasks, including restore operations at the operating system, file systems and/or any application level.
•Managing backup storage capacity and consumption that may include, but is not limited to:
activity reporting, ordering additional storage capacity via My VMwareTM and deleting any backup data in inventory to free up space.

FireEye All backup and restore procedures are in compliance with SOC 2 specifications. Data and server images are backed up digitally and replicated to datacenters in geographically different locations.

VirtueStream Virtustream performs backups on a nightly basis and replicated offsite for further protection. The retention policies are determined by State of Utah per landscape and will be documented during the on-boarding process.

Virtustream provides OS level files are protected with a File System backup and then protect the application with an integrated solution. Some of the databases would dump to a flat file format and those files would be protected with the standard file system backup. b. Method of server image backups

CA APM N/A
MAA Automation tools are used to manage the configuration of MAA internal servers, and can be used for recovery.

CA Agile Systems are not backed up as we have determined rebuilding is faster than restoring.

We manage systems through Chef and the configuration cookbooks are backed up.

ASM Automation tools are used to manage the configuration of ASM internal servers, and can be used for recovery.

Virtru All images are backed up onto cloud storage on two separate provider systems Salesforce Each server has been allocated a different volume group so it can fail over to its backup server within the instance independently of the others.

ServiceNow Please see our response to 8.8.2a
QTS Your data storage requirements are growing exponentially and a scalable comprehensive solution for storage and data protection is absolutely necessary in today’s marketplace. Implementing and managing this necessary storage infrastructure to meet both your application and compliance requirements can be extremely time consuming and costly.

Outsource your storage needs to QTS. With our portfolio of high-performance and cost-effective managed storage services, your data will be secure and guaranteed to be available when you need it. By leveraging market leading hardware and software, coupled with certified processes and personnel, QTS delivers dependable, auditable solutions. #15..
- 854 -
SOLICITATION # CH16012
116
Managed SAN (Shared and Dedicated)- a fast, reliable Fibre Channel (FC) and iSCSI SAN attached storage solution that easily scales to meet your business requirements.

SAN-to-SAN Replication - provides block level asynchronous replication of your Managed SAN FC service between two of our state-of-the-art data centers for your disaster recovery needs.

SAP Ariba See above
Fieldglass Fieldglass does not run server image backups. All systems are designed to be created through a series of templates and scripts.

Hanna The defined backup package includes a file system, operating system files, and the database backup. The frequency in which the backup is executed varies for each of the components mentioned above. This depends on the standard HEC service schedules and the service levels agreed on with the customer.

As part of this service, VMware will:
•Implement and maintain central service components (backup software appliances, backup and archival storage media and associated network topologies) needed to support Data Protection features.
•Perform routine configuration, maintenance and optimization services on behalf of the Data Protection environment and in conformance with industry best practices. •Allocate requisite backup storage based on capacity selections made at the time of subscription enrollment.
•Guarantee storage locality per geographical region for all backup data. •Provide necessary Data Protection service reporting as it is requested.

Participating entities will be responsible for:
•Subscribing to Data Protection as an add-on feature via My VMware and selecting an amount of backup storage capacity commensurate with your requirements. •Creating custom backup protection policies that may include, but are not limited to: affinity settings per VDC, scheduling, and retention periods.
•Registering and unregistering individual vApps and their virtual machine members for scheduled backups using Data Protection.
#15..
- 855 -
SOLICITATION # CH16012
117
•Performing any on-demand backups per vApp and its virtual machine members. •Performing in-place or out-of-place restores per vApp and/or individual virtual machine. •Managing any in-guest recovery tasks, including restore operations at the operating system, file systems and/or any application level.
•Managing backup storage capacity and consumption that may include, but is not limited to:
activity reporting, ordering additional storage capacity via My VMwareTM and deleting any backup data in inventory to free up space.

VirtueStream OS level files are protected with a File System backup and then protect the application with an integrated solution where available. Otherwise databases would dump to a flat file format and those files would be protected with the standard file system backup. Additionally, core microVM’s are automatically replicated to a secondary datacenter for DR purposes. c. Digital location of backup storage (secondary storage, tape, etc.) CA APM Backups are securely replicated to an alternate location (within the same geographic location (e.g. N. America))
MAA Daily differential and full backups reside in the primacy site with weekly data backups residing offsite.

CA Agile Backups are stored in our warm data center.

ASM CA has an alternate site approximately 900 miles from our primary site.

Google Tape
Virtru All backup storage is hosten in two separate cloud storage regions Salesforce Backups are performed daily at each data center facility without stopping access to the application. Backup cloning is transmitted over an encrypted network (our MPLS network across all data centers). Tapes never leave our secure data center facilities, unless they are to be retired and destroyed through a secure destruction process.

ServiceNow Please see our response to 8.8.2a
QTS Managed Tape Rotation and Off-site Storage
This service provides off-site storage for data backup, and helps many customers who must store data separately from a primary IT site to comply with various laws and regulations. QTS has contracted Iron Mountain — a nationally recognized leader in off-site data storage services — to move the data storage media (tapes) to and from a secure, off-site location. The service enables data security, integrity, and restoration in the event of an outage. Off-site data storage via tape ensures that a customer’s information is securely stored in multiple locations. In the event of an emergency,
QTS and Iron Mountain deliver the customer’s media back to their IT environment or wherever they request delivery.

Secure: QTS staff adheres to storage-industry best practices and security procedures. QTS facilities have earned SSAE16-SOC1 accreditations. All media is maintained in a secure environment and locked in secure media containers during transition from the QTS environment to the off-site storage location.

Compliance and Tracking: QTS’s partnership with Iron Mountain provides the ability to offer logging and tracking information about the customer’s off-site media via the SecureSync Web portal.

SAP Ariba See above
#15..
- 856 -
SOLICITATION # CH16012
118
Fieldglass Fieldglass houses all backup data vaults within its production data centers.

Replicas of the vault are performed continuously across data centers more than 1,000 miles apart.

SuccessFactors All data is backed up to the secondary site and can be restored from there in the event of a disaster recovery event at the primary site.

VMware VMware vCloud Air Data Protection provides backup storage to disk FireEye All backup and restore procedures are in compliance with SOC 2 specifications. Data and server images are backed up digitally and replicated to datacenters in geographically different locations.

Backup data is stored on local enterprise class disk storage systems and can be replicated to systems in a secondary Virtustream datacenter. All of Virtustream backup is disk based solution.

Encryption keys are generated at the time of backup application install on the Client VM(s).

Encryption keys are at least AES-128 but can be AES-256. Keys are stored on the backup system databases using the encryption key management tool, the data about the keys on the database is in a unreadable format and cannot be decrypted
d. Alternate data center strategies for primary data centers within the continental United States.

CA APM The location where data is stored is generally in the country where the contract is executed.

MAA The MAA service is currently available from one data center only.

CA Agile We have a hot/warm data configuration with both data centers located within the US but ~1300 miles apart.

ASM ASM has a warm standby with near time data replication between the data centers.

Google Google does not use the N+1 data center assignment due to the risk of cascading failures.

Google's infrastructure makes a dynamic decision for each unique login session which data center is the closest, most highly available data center and which data centers will be used as secondary and tertiary. This assignment would shift for that session throughout the day based on availability and is done completely seamless to the end user.

Virtru We utilize multiple availability zones in AWS, and backup all data stored in our database to separate datacenters
Salesforce Customer Data for customers in Salesforce’s Government Cloud is stored in two of our U.S. data center locations.

The Salesforce service performs near real-time replication at each data center and annual disaster recovery tests for the service verify the projected recovery times and data replication between the production data center and the disaster recovery center. The disaster recovery site is a 100% replica of the primary production site of capacity (host, network, storage, data). Data is transmitted between the primary and disaster recovery data centers across encrypted links. Additionally, backups of data are performed and data is retained on backups at the geographically separated disaster recovery data center location.

ServiceNow Please see our response to 8.8.2a
QTS The QTS cloud alternate storage sites are located at the following data centers: Richmond, VA, Atlanta, GA, and Suwanee, GA.

The entire QTS cloud is a unified standalone Active cloud architecture and is backed up in both Atlanta and Richmond to include:

Customer data in the Service Delivery Hypervisor Cluster, and; Cloud Hypervisor Management Infrastructure and Configuration backups. #15..
- 857 -
SOLICITATION # CH16012
119
The QTS cloud Security documentation is backed up in alternate site located Suwanee, GA to include:

Software/operating Systems, Configuration Data and Security Documentation QTS cloud Management Infrastructure and Configuration Data:

Richmond, VA: The Richmond Cloud Management Layer (Cloud Hypervisor Management Cluster) and Customer data in the Service Delivery Hypervisor Cluster, and Configuration data is backed and staged locally in Richmond on the management layer then backed up across the Site-to-Site Management VPN to Atlanta, GA. This process is separate and independent of the Customers Service Delivery Layer. Atlanta, GA: The Atlanta Cloud Management Layer (Cloud Hypervisor Management Cluster) and Configuration data is backed and staged locally in Atlanta on the management layer then backed up across the Site-to-Site Management VPN to Richmond, VA.

This process is separate and independent of the Customers Service Delivery Layer.

This process maintains both local and remote management layer data and configuration backup for both locations.

Suwanee, GA: software/operating systems, configuration data and security documentation at Rocstor Rocsecure Amphibious X7s with Secure-Encrypted Two-Factor Authentication, Utilizing AES-256 CBC Real-Time Hardware Encryption, with NIST and FIPS 140-2 Certified Cryptographic Modules, in a firerated safe located at Suwanee, GA.

SAP Ariba We have a warm site for failover. Equinix (Our primary data center) is located in San Jose, California and our disaster recovery site is located at CenturyLink Data Center in Sterling, Virginia. The act of failing over from the main data center to the recovery site data center has a design goal recovery time objective (RTO) of four hours. The design goal Recovery Point Objective (RPO) is five minutes.

In the event of a fail-over to the disaster recovery site location, all URLs that customers use to reach the San Jose data center will continue to work. We will notify customers via their email addresses in the event of unplanned downtime.

Internally, we use a documented system recovery plan that outlines the approach and steps for recovering the applications.

The document defines roles and responsibilities in the event of disaster:

Local staff maintains the hardware remotely
We maintain the application software
Processes are in place to keep database and fileservers in sync between primary and backup data centers
In the event of a catastrophe, we will declare the primary data center ""down"" and our local staff will follow a script to start the applications at the remote data center
We test power outage backup scenarios and the Disaster Recovery Plan on a periodic basis.

Fieldglass Please see the answer to Question 8.8.2A. above. Fieldglass' backup data center is located in San Jose, CA.

For example, site A &B may be Santa Clara, CA and Sterling, VA, respectively.

Hybris In the case of catastrophic failure to your primary production data center, we offer anywhere from best efforts to enhanced DR solutions. Enhanced DR #15..
- 858 -
SOLICITATION # CH16012
120
options translates to either RTO of 8 hours and RPO of 1 hour for a warm standby solution, or RTO of 30 minutes to RPO of 30 minutes for a hot-site.

SuccessFactors Each client has a designated primary and backup data center which is geographically separated as well as existing on different power and telecommunications grids.

VMware Customers can purchase Rackware to provide geographical HA and COOP, but this service is not provided as a core capability of vCA or vCGS. VMware AirWatch AirWatch backs up the production environment from the primary US data center to the secondary US data center. Due to FOIA restrictions and the competitive nature of the EMM landscape, we cannot provide additional backup scheduling or image security procedures.

VirtueStream State of Utah can choose any of our FedRAMP Data Centers to host their applications and the DR location. For illustrative purposes, we have assumed the primary data center will be USDC02 and DR data center would be USDC01.

The State of Utah solution has been designed with all workloads running in our San Francisco (USDC02) data center. In the unlikely event of a disaster at the USDC02 facility, State of Utah’s production services will be brought back online in the secondary data center located in Northern Virginia (USDC01). The target Recovery Point Objective (RPO) is 30 minutes for virtual machines with a 2 hour Recovery Time Objective (RTO). All production workloads in the primary node will have reserved compute capacity in the secondary node when deployed using Enterprise Reserve µVMs and replicated storage. All virtual machine disaster recovery services are based on storage replication. The primary and secondary data centers are interconnected by diverse private 10 Gbps Sonet rings from separate Tier 1 providers – AboveNet and Level 3. Data is replicated asynchronously and continuously between the two data centers. Virtustream also performs backups on a nightly basis and replicated offsite for further protection. The retention policies are determined by State of Utah per landscape and will be documented during the on-boarding process.

8.9 Data Protection

8.9.1 Specify standard encryption technologies and options to protect sensitive data, depending on the particular service model that you intend to provide under this Master Agreement, while in transit or at rest.

CA APM All data is encrypted via TLS Mutual Auth during transit #15..
- 859 -
SOLICITATION # CH16012
121
MAA HTTPS/TLS encrypts the data in-transit. Sensitive data rest is encrypted using native encryption of SQL and NoSQL vendors.

CA Agile All data in transit is encrypted - we support TLS 1+. For data at rest we have both database and disk level encryption. DB encryption utilizes Oracle TDE w/ AES-256.

ASM All data transmitted between data centers is encrypted in-transit. HTTPS/TLS is used by the ASM dashboard and API.

All data in transit between Google's Data Centers traverses across Google's private fiber network using a customized, proprietary encryption technology.

Google hard drives leverage technologies like FDE (full disk encryption) and drive locking, to protect data at rest.

These methods of encryption are fully managed by Google and Google's Key Management Servers based on 128-bit or stronger Advanced Encryption Standard (AES).

Encryption Keys and Ciphers Supported by Google
Protocols
TLS 1.2
TLS 1.1
TLS 1.0
SSL 3.04
QUIC
Cipher suites
ECDHE_RSA with AES
ECDHE_RSA with 3DES
ECDHE_ECDSA
RSA with AES
RSA with 3DES
Signing keys
RSA 2048
ECDSA P-256
Hash functions
SHA384
SHA256
SHA1
MD5
AODocs All data in transit is encrypted via SSL/TLS
Vendor interfaces are available only via HTTPS
Email is exchanged via SMTP/TLS
All AODocs data are stored in Google AppEngine.

Virtru AES-256 in CBC mode, TLS 1.2 ECDHE
Salesforce Encryption Capabilities
Salesforce has many customers that are subject to laws pertaining to the processing of personally identifiable information (PII) or personal data. Salesforce offers its customers a broad spectrum of functionalities and customer-controlled security features that its customers may implement in their #15..
- 860 -
SOLICITATION # CH16012
122
respective uses of the Salesforce services. Salesforce believes that these provide its customers the flexibility to comply with laws with stringent privacy and security requirements.

Data In Motion
All transmissions between the user and the Salesforce Services are TLS encrypted with a 2048-bit Public Key. The Services use International/Global Step Up TLS certificates, with AES 256-bit encryption by default.

Data At Rest
Salesforce includes a feature to encrypt custom text fields (ECF):

The fields can be masked appropriately for specific data types (i.e., credit card number, Social Security Number, National Insurance Number, Social Insurance Number).

Access to read the masked parts of the fields is limited by the ""View Encrypted Data"" permission, which is not enabled by default.

Customers can manage their encryption key based on their organization’s security needs and regulatory requirements.

Encrypted fields are encrypted with 128-bit keys and use the AES (Advanced Encryption Standard) algorithm.

Additional Salesforce Encryption Capabilities
Apex Code extends the powerful and proven success of the Force.com platform by introducing the ability to write code that runs on Salesforce servers. This language makes possible the development of a new class of application and features deployed entirely on demand. Using Apex, your Agency can create user interface classes that utilize the Apex crypto class to encrypt field level data up to AES 256-bit encryption.

Third Party Encryption Solutions (Additional License Option) Should additional encryption be required, third party solutions such as CipherCloud, Skyhigh, and PerspecSys are available on the Salesforce AppExchange. These solutions offer data loss prevention, user activity and monitoring, malware protection, as well as data protection with encryption and tokenization. Data can be encrypted and masked at rest, keys managed and stored in Salesforce, and compliance controls to prevent unauthorized access to data and keys.

ServiceNow ServiceNow makes use of encryption for both data in transit and data at rest. ServiceNow provides optional capabilities with regard to the encryption of data at rest within the system, which customers can apply at their own discretion.

Encryption in Transit
ServiceNow customers access their instances over the Internet using forced Transport Layer Security (TLS) encryption (AES128/256) for all user access. The level of encryption is based on the browser and must be configured by the customer as ServiceNow does not modify any browser settings. All attempts to access ServiceNow over HTTP are redirected to HTTPS.

Integration Encryption
For integrations such as LDAP and Web Services, ServiceNow provides customers with the ability to encrypt traffic. LDAP can be configured to run over SSL, this requires customers to provide a certificate for the specific LDAP server. Certificates may also be stored within an instance to allow encrypted transmission for Web Services integrations. FTPS and SCP can be used as file transfer methods to securely transfer files to ServiceNow.

Email Encryption
Customers may configure their instance to generate emails to their users from the instance.

ServiceNow provides the capability to receive email over TLS. Customers are able to configure their #15..
- 861 -
SOLICITATION # CH16012
123
email system to send email to ServiceNow over TLS and ServiceNow will receive that email over TLS.

Encryption at Rest
ServiceNow can provide three types of encryption for data at rest that are implemented by the customer or by customer request in the case of edge encryption and dedicated hardware. •Column encryption of customer added fields and attachments: Provides data encryption using AES128/256 or 3DES symmetric key encryption. The customer provides the keys for this encryption.

Data stored in these fields cannot be searched or reported on and this does not support out of the box fields.
•Edge Encryption Proxy: With optional additional cost Edge Encryption, the customers create and control their encryption keys within their own network. Edge Encryption includes a proxy application that resides in a customer's network. This encrypts data before it is sent (also encrypted in transit) from the customer’s environment to the ServiceNow instance. The data always remains encrypted whilst stored in the instance and the data along with the keys and the encryption configuration is never accessible by ServiceNow. Requests for encrypted data must also be made through the proxy application and is therefore decrypted only within the customer’s network before being sent to their end users client browser. Please note that data encrypted with Edge encryption cannot be used by back-end scripts or processes and searching and reporting behavior is also modified through its use. •Full disk encryption: Provides via self-encrypting hard drives with AES256 bit encryption. This encryption capability is only available through the purchase of dedicated ServiceNow hardware at an additional cost. This delivers “at-rest” protection only and is focused solely on preventing data exposure through the loss or theft of hard disks holding customer data.

Wherever possible, ServiceNow leverages existing FIPS 140-2 certified technologies.

Docusign Secure, private SSL 256 bit viewing session
Anti-tampering controls
Signature verification of signing events
Unalterable, systematic capture of signing data
Digital certificate technology
Customer configurable data retention program
SAP Ariba We enforce minimum AES 128-bit encryption using Transport Layer Security (TLS) for all sessions. We encrypt only PCI-DSS data in the database where appropriate and in support of PCI-DSS compliance. All backup media is encrypted using AES 256-bit cipher prior to transporting to off-site storage Fieldglass Our application provides a high level of application-level security through a combination of encryption, page-level access checking, document envelopes, and activity logging. Application security is handled through a combination of programming checks, application server configuration, and database server configuration. Fieldglass uses 2048-bit SSL (HTTPS) encryption for all data transmissions over the public Internet, including data shared between the product and end users and data shared between the product and back-end systems. Data is decrypted on the internal VLAN to allow IDS monitoring.

Passwords are hashed using a one-way hash based on the SHA-256 encryption algorithm. The hash value is saved within the database; not the password.

The base Fieldglass application does not require data that would require breach notifications if compromised. While customers should not store sensitive personal data, custom fields may be defined by the customer that can be encrypted with AES-256 to capture any pertinent and permissible data points. These fields can also be optionally masked from view while entering and viewing the fields in the application.
#15..
- 862 -
SOLICITATION # CH16012
124
Hanna Initial encrypted data load - physical encrypted data transfer possible Encrypted connection to HEC DC - SAP solutions support various VPN-device vendors and specifications that can be used to setup a secure and encrypted tunnel between cloud solutions from SAP on the customer network segment and the customer site.

Optional Data at Rest encryption - HANA encryption at Rest activated during system build upon request.

Moreover data at rest is properly protected against unauthorized access, these controls include
•Physical security controls in SAP HEC Datacenters in place •Physical separation of “online” storage and backup (physically separated DC sections)
•Strict User and Access Management (segregation of duties) •Authorization Management according to the need to know principle also for administrative accounts, e.g. DBAs have limited authorizations to perform DB operation only
•Security Logging and Security Monitoring for critical activities / access Optional encryption in transit can be configured on customer request which requires a different subset of DC's from delivery perspective.

Hybris Encryption is provided through SSL connectivity at the web application layer as well as VPN secure encrypted tunnel for back-office integration connectivity. Further encryption capabilities are available through the hybris Advanced Security Module (optional).

SuccessFactors All data in the system is treated as highly sensitive information. The application and infrastructure offers Strong Encryption, Encryption in Transit and Encryption at Rest, including the following (some of these are options that are not included in the base services):

Transport Layer Security (TLS) technology, consisting of a public key and a private key, to protect sensitive information. The public key is used to encrypt information and the private key is used to decipher it. When a Web browser points to a secured domain, a handshake authenticates the server (website) and the client (browser). An encryption method is established with a unique session key. Customers may then begin a secure session that facilitates message privacy and integrity.

Hitachi SAN storage arrays provide a reliable safe, secure data storage environment. We use AES-256 bit encryption to secure data at the block level of our storage systems.

Backups are performed ""disk to disk"". Data is transported over 3DES VPN and stored on encrypted disk using Data Domain replication technology VMware VMWare IaaS Services
Data in transit will be encrypted based upon the customers selected method of transport (SSL or IPSEC VPN).

Data at rest encryption is the responsibility of the customer. By example HyTrust is an endpoint encryption choice for many vCloud Air customers.

VMWare AirWatch
AirWatch leverages strong, non-proprietary encryption algorithms to protect applicable data at rest and in transit. Due to FOIA restrictions and the competitive nature of the EMM landscape, we cannot provide specific encryption algorithms and protocols.

FireEye Each of the FireEye offered solutions have their own implementations and practices concerning encryption:
#15..
- 863 -
SOLICITATION # CH16012
125

1.Email Threat Prevention (ETP)
ETP offers TLS encryption support for data in transmission. ETP only stores email found to contain malware or malicious content and does not encrypt this data.

2.Mobile Threat Prevention (MTP)
All data in transit is protected using standard TLS encryption. The database itself is not encrypted.

However, all systems are under strict security and access control rules, in compliant with SOC 2 regulations.

3.Threat Analytics Platform (TAP)
All data in-transit between a customer environment and TAP instance in Amazon AWS is encrypted with a 256-bit Twofish key. No customer data is stored on disk, however data stores leverage Amazon AWS S3 encryption settings.

4.FireEye as a Service (FaaS CV)
FireEye appliance communications, and FireEye as a Service (FaaS) non-person entity (NPE) secured inter-process communications employs industry standard HTTPS encrypted web interfaces, and multifactor SSH management access for the end-to-end protection of sensitive customer data. In addition, FaaS communicates with FireEye appliances with additional layers of protected symmetric key exchanges protecting and establishing encrypted channels for inter-device information.

VirtueStream Encryption keys are generated at the time of backup application install on the Client VM(s).

Encryption keys are at least AES-128 but can be AES-256. Keys are stored on the backup system databases using the encryption key management tool. The data about the keys on the database is in an unreadable format and cannot be decrypted.

8.9.2 Describe whether or not it is willing to sign relevant and applicable Business Associate Agreement or any other agreement that may be necessary to protect data with a Purchasing Entity.

Carahsoft would be willing to sign Business Associate Agreements or similar agreements that are set forth with a purpose of protecting customer data.

8.9.3 Offeror must describe how it will only use data for purposes defined in the Master Agreement, participating addendum, or related service level agreement. Offeror shall not use the government data or government related data for any other purpose including but not limited to data mining. Offeror or its subcontractors shall not resell nor otherwise redistribute information gained from its access to the data received as a result of this RFP.

Agreed and understood. Carahsoft and its’ subcontractors will not use any government data for purposes other than those outlined within the frameworks of this contract.

8.10 Service Level Agreements

8.10.1 Offeror must describe whether your sample Service Level Agreement is negotiable. If not describe how it benefits purchasing entity’s not to negotiate your Service Level Agreement.

CA APM The target availability SLA of 99.8% is standard and not negotiable MAA The target availability SLA of 99.8% is standard and not negotiable CA Agile The target availability SLA of 99.8% is standard and not negotiable ASM The target availability SLA of 99.8% is standard and not negotiable Google Google maintains a single SLA for all customers.

AODocs AODocs maintains a single SLA for all customers.

Virtru Agreement can be negotiated for certain large customers. #15..
- 864 -
SOLICITATION # CH16012
126
Salesforce Salesforce does not typically offer Service Level Agreements as part of the base service offering.

Our approach is to offer a service with high availability and fast resolution of problems. If a customer requires an SLA it will be negotiated separately.

Salesforce has maintained high levels of availability across all Salesforce instances since inception. As the only on-demand vendor to provide daily service-quality data on a public Web site (http://trust.salesforce.com), Salesforce proves that we are the leader in availability. And by making its track record completely transparent, Salesforce proves we are worthy of our customers’ trust. To ensure maximum uptime and continuous availability, Salesforce provides the best redundant data protection and most advanced facilities protection available, along with a complete data recovery plan—all without affecting performance.

ServiceNow ServiceNow’s SLA is not negotiable. ServiceNow delivers the same level of world-class support to all customers as described in the “Subscription Service Guide” included with this response.

ServiceNow’s homogeneous private cloud environment where all applications are on a single platform offers ServiceNow a competitive advantage in being able to concentrate its efforts to make the customer’s user experience the best possible.

The ServiceNow environment is a private cloud, fully owned and operated by ServiceNow.

ServiceNow’s experience & understanding of its private cloud and the benefits provided by a consistent infrastructure and standardized processes allow it to provide a high level of security, availability, and performance in a cost effective and reliable manner.

Docusign Yes SLA's can be modified per contractual agreement.

SAP Ariba Our SLAs are not negotiable. As mentioned above, the SAP Ariba Cloud Service Level Agreement includes a 99.5% uptime.

Fieldglass Our SLA is designed to (i) meet the expectations of our customers and (ii) fit with our operational protocols and capabilities. Because of this our ability to make significant changes to the SLA is limited. However, we are always willing to work with a customer to try and determine how our SLA can address a particular business.

Hanna Our SLAs are not negotiable for Support. Under our SLA, System Availability is 99.5% during each month for productive systems. Customers may claim a credit of 2% of Monthly Subscription Fees for each 1% below SLA (not to exceed 100% of Monthly Subscription Fees). This credit may be applied to a future invoice relating to Cloud Service that did not meet the System Availability SLA.

Hybris Our SLAs are not negotiable for Support. Under our SLA, System Availability is 99.5% during each month for productive systems. Customers may claim a credit of 2% of Monthly Subscription Fees for each 1% below SLA (not to exceed 100% of Monthly Subscription Fees). This credit may be applied to a #15..
- 865 -
SOLICITATION # CH16012
127
future invoice relating to Cloud Service that did not meet the System Availability SLA.

SuccessFactors Our SLAs are not negotiable for Support. Under our SLA, System Availability is 99.5% during each month for productive systems. Customers may claim a credit of 2% of Monthly Subscription Fees for each 1% below SLA (not to exceed 100% of Monthly Subscription Fees). This credit may be applied to a future invoice relating to Cloud Service that did not meet the System Availability SLA.

VMware VMware’s SLAs are specific to the product or service, and are subject to the product’s Terms of Service (which are also provided). The SLA is specific to the product rather than to the customer, to define the expected availability and recourse should the product not meet the stated availability levels. VMware’s existing customers include Federal, State and Local government customers as well as commercial customers, who successfully utilize our products in accordance with the current SLA and Terms of Service and their applicable, mandatory laws and regulations.

Accordingly, VMware is confident that the existing SLA meets the state and local government requirements, as it is being utilized currently.

VMware IaaS Services
To best work with our customers and improve our terms in alignment with market requirements, VMware is willing to review the terms of its SLA during negotiations of the Participating Addendum to identify how the existing SLA meets the state’s applicable and mandatory laws and regulations.

If it becomes apparent to the parties that it is necessary to revise VMware’s SLA to comply with the state’s applicable and mandatory laws, VMware will enter into discussions about how to best address these requirements
Vmware SaaS Services
VMware AirWatch
AirWatch Service Level Agreements are non-negotiable. AirWatch has a guaranteed uptime SLA of 99.9%. Please refer to the AirWatch Hosted Services Policy for additional information. VMware AirWatch acknowledges that if it is awarded a contract under the RFP that it will annually certify to the Lead State that it still meets or exceeds the mandatory minimum requirements and technical specifications of the RFP.

FireEye SLAs are negotiable.

VirtueStream Virtustream’s Service Level Agreement is generally not negotiable. Our SLAs are one of the strictest in the market place. To change the SLA would mean either changing the design or create operational disruption and higher cost by engaging in non-standard processes and practices. For example, Virtustream’s SLA for Tier 1 storage is 10 ms latency. For Virtustream to change this latency means changing the design for Tier 1 storage.

8.10.2 Offeror, as part of its proposal, must provide a sample of its Service Level Agreement, which should define the performance and other operating parameters within which the infrastructure must operate to meet IT System and Purchasing Entity’s requirements.

Below is a sample of Virtustream’s Service Level Agreement:

SERVICE LEVEL FRAMEWORK
The service levels (“Service Levels”) applicable to the Services specified in Sections 1 and 2 are set forth in Schedule B to this SD (“Service Levels for Cloud Platform Services”). The framework that governs all Service Levels is set forth in this Section.

Commencement of Service Levels
Commencing thirty (30) days from the Service Start Date (as set forth in the applicable Order Form), Virtustream’s performance of the Services will meet each applicable Service Level. If Virtustream’s performance of the Services #15..
- 866 -
SOLICITATION # CH16012
128
does not meet the applicable Service Level, then Virtustream will use commercially reasonable efforts to restore its performance to meet such Service Level.

Service Level Reports
Service Levels will be calculated and measured monthly by Virtustream on a calendar month basis and reported each month for the previous month. The reports will be provided to Customer by the tenth (10th) working day of the month following that to which such report relates, commencing on the second (2nd) month following the Service Start Date and each month thereafter. The monthly service level report will contain at least the following items: (i) Uptime statistics for the month concerned; (ii) an analysis of reported incidents over the previous month, broken down by type for discussion; (iii) action plans for items giving rise to concern; (iv) comments and observations on any issues arising from Virtustream’s performance monitoring activities; (v) recommendations on service delivery strategies to maintain or enhance the service level; and (vi) review of general business requirements (“Service Level Report”).

Cloud Platform Services (CPS) has its own specific service levels as described in this document. Cloud Cover Services (CCS) has service levels that pertain to the CCS offerings and are reported separately. Not all Virtustream customers have CCS but all Virtustream customers use CPS.

Service Level Review Meetings
Monthly Service Level review meetings will be conducted by Virtustream with Customer where the monthly Service Level report specified above will be discussed. If any of the Service Levels measured over the previous calendar month period is not achieved in that month, then Virtustream will include the steps taken to rectify the problem in the next monthly Service Level Report. In addition, the issue shall be an agenda topic for discussion at the next monthly service review meeting. Additionally, after restoring service or otherwise resolving any immediate problem as specified in this SD, if Virtustream fails to provide Services in accordance with the Service Levels, Virtustream shall:
a.Promptly investigate and report on the causes of such problem; b.Provide a Root Cause Analysis of such failure as soon as practical after such failure or at Customer’s request; c.Correct such problem that is Virtustream’s fault or responsibility, as soon as reasonably practicable and coordinate the correction of such problem if Virtustream does not have responsibility for the cause of such problem. d.Advise Customer of the status of remedial efforts being undertaken with respect to such problem; e.Demonstrate to Customer’s reasonable satisfaction that the causes of such problem (that is Virtustream’s fault or responsibility) have been or shall be corrected on a permanent basis; and f.Take corrective actions to prevent any recurrence of such problem (that is Virtustream’s fault or responsibility).

Root Cause Analysis
Promptly following Virtustream’s failure to meet a Service Level, Virtustream will perform a root cause analysis to determine the reason for that failure. Upon Virtustream’s determination of the cause of such failure, it will provide to Customer a preliminary report citing the cause of such failure. If Virtustream determines that the failure was due to Virtustream, an additional report will be provided that details the root causes of the failure, and which details any measures that should be taken to minimize the possibility that such failures will re-occur. Virtustream will correct the problem and use reasonable commercial efforts to minimize the re-occurrence of such failures.

Service Level Exceptions
Virtustream shall not be liable for any failure to meet the Service Levels, to the extent such failure was caused by one or more of the following:
a.A failure of Customer or any of its employees, agents or contractors (including any of Customer’s third party service providers) to perform any of its responsibilities under this SD; b.Any act or omission of Customer or any of its employees, agents or contractors (including Customer’s third party service providers or other third parties acting on behalf of Customer); c.Any hardware, software or other product of a third-party or Customer equipment; d.Any failure of Customer to secure the proper access rights or maintenance and support services with respect to any component of the Services (e.g., hardware, software, network, maintenance) for which Virtustream does not bear operational responsibility;
e.Downtimes resulting from a Virtustream’s scheduled maintenance windows; f.Customer’s reprioritization of the tasks to be performed by Virtustream where such reprioritization causes Virtustream to miss a Service Level;
g.Viruses; provided that the infected Virtustream-provided system had virus protection for which the virus protection software updates were up to date;
#15..
- 867 -
SOLICITATION # CH16012
129
h.An election by Customer to purchase a base commitment that is not sufficient to run Customer’s system (e.g., If a customer elects to size a µVM pool that is insufficient to run the designated workload); i.Issues occurring outside of standard working hours (as defined for business level customers) — for which the Service Level Objectives (SLOs) do not apply;
j.Cloud Cover Services (CCS) offerings — for which the Service Level Objectives (SLOs) do not apply; k.Resolution delays due to lack of client response and/or Customer provided credential based information; l.Priority levels not agreed upon by both customer and supplier; m.Claims of performance degradation not substantiated through Customer provided diagnostic testing results.

8.11 Data Disposal
Specify your data disposal procedures and policies and destruction confirmation process.

The disposal of data will vary based on the type of service that is being provided to the Purchasing Entity by Salesforce. As an example, please see Salesforce response for disposal of government data: Data Disposal and Destruction
In the event of termination of the Salesforce service, requests by your Agency made within 30 days after the effective date of termination or expiration of the Subscription Agreement, Salesforce will make your data available to you for export or download. Once the export has been completed, an email will be sent to you containing a link where you can download a .zip file that contains multiple .csv (spreadsheets) files, each representing your Salesforce objects. Your data on disk is flagged within the database and set to inactive status or what can be referred to as a ""soft delete."" This data is no longer available or accessible to the application but is backed up in the full database backup process. The data remains in this state for 180 days; this is done in the event that the customer decides to resume services or needs the data for a legal reason. At 180 days, the data is marked for deletion (""hard delete"") and will be deleted after 30 more days. Once this ""hard delete"" is executed the customer data is physically deleted and non-recoverable from the database. Following the purge, the data will remain on backup for an additional 90 days prior to being overwritten and unrecoverable.

Media Sanitization
Salesforce has an established process for sanitizing media consistent with industry guidelines and consistent with NIST SP 800-88 Guidelines for Media Sanitization [MP-6].

8.12 Performance Measures and Reporting

8.12.1 Describe your ability to guarantee reliability and uptime greater than 99.5%. Additional points will be awarded for 99.9% or greater availability.

CA APM CA SaaS production environments have an SLA at 99.8% uptime calculated arrear monthly. 1hr response for P1 issues.

MAA CA MAA SaaS production environments have an SLA at 99.8% uptime calculated arrear monthly. 1hr response for P1 issues.

CA Agile For Rally SaaS Unlimited Edition customers, our goal is to provide 99.9% up-time during each calendar quarter. If in any calendar quarter an up-time of 99.5% is not met and our customers were negatively impacted (for example, were unable to login to Rally), we will provide a service credit equal to one month of fees for use of the Rally Service.

ASM CA SaaS production environments have an SLA at 99.8% uptime calculated arrear monthly. 1hr response for P1 issues.
#15..
- 868 -
SOLICITATION # CH16012
130
Google Google Apps services will be operational and available to the customer at least 99.9% of the time in any calendar month. If Google does not meet the Google Apps SLA, the customer will be eligible to receive service credits.

AODocs Altirnao agrees to use commercially reasonable efforts to meet or exceed the Availability Service Level. The Availability Service Level means the Service is available for use by Users for not less than 99.5% of the time excluding the Exclusion Events (as defined below).

Virtru Virtru provides the same SLA as Gmail
Salesforce Salesforce has maintained high levels of availability across all Salesforce instances since inception. As the only on-demand vendor to provide daily service-quality data on a public Web site (http://trust.salesforce.com), Salesforce proves that we are the leader in availability. And by making its track record completely transparent, Salesforce proves we are worthy of our customers’ trust. To ensure maximum uptime and continuous availability, Salesforce provides the best redundant data protection and most advanced facilities protection available, along with a complete data recovery plan—all without affecting performance.

Salesforce does not typically offer Service Level Agreements as part of the base service offering.

Our approach is to offer a service with high availability and fast resolution of problems. If a customer requires an SLA it will be negotiated separately.

ServiceNow ServiceNow’s Availablity SLA Is 99.8%, excluding Excused Downtime of up to two hours per month. This is achievable through the use of the Advanced High Availability (AHA) architecture described in 8.8.1 that provides failover capabilities for maintenance and service disruptions to minimize downtime.

Docusign DocuSign offers carrier class bank grade, “always on” availability. This means DocuSign has eliminated monthly maintenance (taking 0 minutes of planned downtime per year). We are a multiyear 99.99% operation. Please view the DocuSign Trust site (trust.docusign.com) for additional information.

QTS TIA 942 requirements for the design, construction, and management of Tier III facilities is the design basis for all QTS data centers. Using best-of-breed technologies, QTS’s highly trained, certified data center professionals tailor efficient and reliable solutions to meet individual customer requirements. Primary commercial power is delivered to facilities through diverse, redundant paths to eliminate single points of failure between the nearest electrical substation (usually on-site) and each piece of customer IT equipment. QTS allows users to inspect the engineering drawings to validate that truly redundant N+1 architectures are in place and that the 99.999% uptime guarantee is supported by sound engineering and operational policies. Meaningful Service Level Agreements are backed by financial penalties spelled out in tailored Master Service Agreements with each customer.

SAP Ariba Our data centers have 99.99% uptime. Our application maintains a 99.% official SLA with the last 12 months seeing 99.98% uptime.

Fieldglass For customers hosted in US data centers, we commit to a 99% availability standard. For customers who choose to be hosted in EU data centers, our availability standard is 96.7%. If we fail to meet the availability standard, the #15..
- 869 -
SOLICITATION # CH16012
131
customer may be entitled to receive a credit equal to two percent (2%) of its transaction fees for the service for that month for each one percent (1%) (or portion thereof) by which we fail to achieve such level, up to one hundred percent (100%) of the fees for such month.

Hanna The SAP HEC offers a 99.5% uptime availability SLA for production (PRD) instances and a 95% uptime availability for quality assurance (QAS) and development (DEV) instances. This means that after taking in to account the monthly maintenance window, the TOTAL solution will be available 99.5% of the time - not that we expect it, but this equates to about 3 hours of unscheduled downtime per month. The concept of "total solution" is important because the HEC provides Holistic Solution Availability - our single SLA covers the FULL SOLUTION STACK which means it covers the Infrastructure, Operating System, local DB, the HANA in-Memory DB, and the Application layers. We do not offer individual SLAs on individual layers because if one layer fails, the entire solution is down Hybris The environment is completely redundant and based on a high availability architecture. The hybris private cloud is based on VMWare virtualization and is fully load-balanced.

Customer systems reside on a redundant server infrastructure ensuring that if one server (or part of the server infrastructure) fails, a backup is in place to ensure that systems remain operational.

The infrastructure is load-balanced ensuring high availability of the system during peak usage periods. The load-balancing allows the systems to support large volumes of simultaneous users balancing the load of activity across multiple servers ensuring peak performance of all systems.

Redundancy of Infrastructure Components:
• Multihome connectivity to the Internet backbone. Multiple connections to the internet, allows optimizing the best route as well as redundancy. • At an individual component level, all devices have dual physical power connections to protect against any individual power supply failure • Redundant Firewalls
• Redundant Routers,
• Redundant Load balancers
• Virtualized Application and Web servers
• Clustered Database - active active cluster mode for enhanced redundancy • High performance SAN storage- from Netapp, with built in redundancy at the disk level (NetApp proprietary RAID technology, similar to a RAID 5) • The data centers provide diverse power (Uninterruptible Power Supply), backup generators and air conditioning systems required to ensure equipment remain online with optimal performance 24/7/365.
• SLA: contractually we provide a guarantee on the performance of your application at 99.9% availability.

SuccessFactors We contractually guarantee a minimum of 99.5% system uptime, with the exception of regularly scheduled and emergency maintenance. #15..
- 870 -
SOLICITATION # CH16012
132
VMware VMware’s SLAs for vCloud Air. Our Availability Commitment ranges from 99.99% - 99.95 based on the Class of Service. The SLAs are subject to the terms of the applicable Terms of Service for the VMware vCloud Air Service Offerings. Dedicated Cloud 99.99%, Virtual Private Cloud 99.95%, Data Protection 99.95%, Disaster Recovery 99.95%, Virtual Private Cloud On-Demand 99.95% Availability Commitment.

IaaS Object Storage: Object Storage powered by Google CLoud Platform (standard) 99.9%, Object Storage powered by Google (durable reduced availability) 99.0% and Object Storage (nearline storage) 99%.
vCloud Government Service Dedicated Cloud 99.95% and Virtual Private Cloud 99.9% Availability Commitment.

FireEye FireEye undertakes all reasonable efforts to ensure the availability for 99.9% of the time during each calendar month. In the event the solution does not meet the monthly service availability, FireEye will provide a credit as outlined in the SLA.

VirtueStream Virtustream solution for VM workload includes the uptime of 99.99%.

8.12.2 Provide your standard uptime service and related Service Level Agreement (SLA) criteria.

CA APM Uptime SLA target is 99.8%
CA runs test scripts using application monitoring tools on the Production system to verify that the CA SaaS service is available. Test scripts are run approximately once every ten (10) minutes, twenty┽four (24) hours per day, seven days per week, throughout the contracted term of the service.

MAA Uptime SLA target is 99.8%
CA runs test scripts using application monitoring tools on the Production system to verify that the CA SaaS service is available. Test scripts are run approximately once every ten (10) minutes, twenty┽four (24) hours per day, seven days per week, throughout the contracted term of the service.

CA Agile See above.

ASM Uptime SLA target is 99.8%
CA runs test scripts using application monitoring tools on the Production system to verify that the CA SaaS service is available. Test scripts are run approximately once every ten (10) minutes, twenty┽four (24) hours per day, seven days per week, throughout the contracted term of the service.

Google Google Cloud Platform SLA: Google's Cloud platform has SLAs ranging from 99.9 - 99.95% uptime.

Salesforce Please see response to 8.12.1 above.

Salesforce does not typically offer Service Level Agreements as part of the base service offering.

Our approach is to offer a service with high availability and fast resolution of problems. If a customer requires an SLA it will be negotiated separately.

ServiceNow DEFINITIONS
(a) “Available” means that the Subscription Service can be accessed by authorized users. #15..
- 871 -
SOLICITATION # CH16012
133
(b) “Excused Downtime” means: (i) Maintenance Time of up to two (2) hours per month; and (ii) any time the Subscription Service is not Available due to circumstances beyond ServiceNow’s control, including without limitation modifications of the Subscription Service by any person other than ServiceNow or a person acting at ServiceNow’s direction, a Force Majeure Event, general Internet outages, failure of Customer’s infrastructure or connectivity (including without limitation, direct connectivity and virtual private network (VPN) connectivity to the Subscription Service), computer and telecommunications failures and delays, and network intrusions or denial-of-service or other criminal attacks.
(c) “Maintenance Time” means the time the Subscription Service is not Available due to service maintenance.
(d) “Availability SLA” means the percentage of total time during which Customer’s production instances of the Subscription Service are Available during a calendar month, excluding Excused Downtime.

AVAILABILITY
If Customer’s production instances of the Subscription Service fall below the Availability SLA of ninety-nine and eight-tenths percent (99.8%) during a calendar month, Customer’s exclusive remedy for failure of the Subscription Service to meet the Availability SLAs is either: (1) to request that the affected Subscription Term be extended for the number of minutes the Subscription Service was not Available in the month in accordance with the Availability SLA; or (2) to request that ServiceNow issue a service credit to Customer for the dollar value of the number of minutes the Subscription Service was not Available in the month in accordance with the Availability SLA (determined at the deemed per minute rate ServiceNow charges to Customer for Customer’s use of the affected Subscription Service), which Customer may request ServiceNow apply to the next invoice for subscription fees.

REQUESTS
Customer must request all service credits or extensions in writing to ServiceNow within thirty (30) days of the end of the month in which the Availability SLA was not met, identifying the support requests relating to the period Customer’s production instances of the Subscription Service was not Available. The total amount of service credits for any month may not exceed the subscription fee for the affected Subscription Service for the month, and has no cash value. ServiceNow may delay issuing service credits until such amounts reach one thousand U.S. dollars ($1,000) or equivalent currency specified in the applicable Order Form.

Docusign Our industry-leading, global support model is there to back you up, no matter where you do your business. We provide you access to the expertise you want, whether through our communities, our knowledge base and on-demand training, or our team of experienced technical support professionals, who know you and your solutions.

QTS 99.999% uptime guarantee. Meaningful Service Level Agreements are backed by financial penalties spelled out in tailored Master Service Agreements with each customer.

SAP Ariba We guarantee a 99.5% system availability percentage during each month for production versions, with the exception of regularly scheduled and emergency maintenance.

Hanna Technical Availability for PRD systems - 99.5%
Technical Availability for non PRD systems - 95%
SAP shall track and report to Customer the “Technical Availability” in a monthly summary report. SAP HEC can offer a contractual SLA of 99.7 on a case by case basis.

Hybris SLA: contractually we provide a guarantee on the performance of your application at 99.9% availability.
#15..
- 872 -
SOLICITATION # CH16012
134
SuccessFactors We guarantee a 99.5% system availability percentage during each month for production versions, with the exception of regularly scheduled and emergency maintenance.

VMware AirWatch Service Level Agreements are non-negotiable. AirWatch has a guaranteed uptime SLA of 99.9%. Please refer to the AirWatch Hosted Services Policy for additional information. VMware AirWatch acknowledges that if it is awarded a contract under the RFP that it will annually certify to the Lead State that it still meets or exceeds the mandatory minimum requirements and technical specifications of the RFP.

FireEye FireEye cloud solutions are available 99.9% of the time in any calendar month, other than mutually agreed upon and scheduled downtime.

VirtueStream Virtustream IaaS Compute is comprised of computing resources hosted in secure data centers that replace the physical computing hardware traditionally housed on Customer site. These resources include physical servers which are logically divided into VMs (virtual machines), each with an allocation of CPU and memory, and linked to storage.

Terms used in the detailed descriptions below:
•µVM. Pronounced “micro VM,” this is Virtustream’s fine-grained unit of measurement designed to accurately measure the actual consumption of cloud resources. A µVM is a unit of computing resources, comprised of CPU, memory, storage IOPS, and associated local network bandwidth.

The usage of each µVM resource component (CPU, memory, storage input/output, and network bandwidth) is measured at five minute intervals — one unit each for 200MHz of CPU, 768MiB of memory, 40 storage fabric input/output operations per second (IOPS), and 2Mbps of local network bandwidth. The highest of the four is averaged per hour, and the hour values averaged across the month to determine the overall µVM usage for the month.

Note: The measurement is performed at the aggregate level — across Customer’s entire µVM resource pool. Bandwidth usage is only within the data center. •Basic Plus uVM. These terms differentiate the two ways Virtustream offers µVMs services. “Basic Plus” µVM services are limited to a single Virtustream data center and have 99.99% availability for the Customer’s committed level.
“Reserve” µVM service is an equivalent quantity of Basic Plus µVMs reserved at Customerdesignated secondary Virtustream data center for on-going operation during disaster events and during scheduled disaster recovery (DR) exercises.

In both cases, overage resources required (“surges”) up to 20% above the committed level from the contract /order form are provided at the same availability level. If Customer experiences consistent overage above this level, then Customer should reset the committed level. •High Memory. Virtustream offers competitive pricing on compute services for applications that require large amounts of memory (64GiB or more).
•Enterprise These terms identify the “zones” within which µVMs are made available for consumption. The “Enterprise” zone is for use by non-Internet facing workloads. Systems deployed into the Enterprise zone are not directly accessible from the Internet.

Enterprise Basic Plus µVM (IC-uVM-BASP-ENT)
Enterprise Basic Plus µVMs reside in a single Virtustream data center, in the Enterprise zone.

Resource availability is 99.99%, and only at the designated data center.

Overage resources required (“surges”) up to 20% above the committed level from the contract /order form is provided at the same availability level. If Customer experiences consistent overage above this level, then Customer should reset the committed level.

Service Level. 99.99% availability for the Customer’s committed level.

Billing. Monthly, based on resource usage.

Tier 0 Block Storage - Local Only (IC-STO-TOA-LOC)
Block storage with a Latency Service Level of 3ms that is maintained in a single data center with no replication capability.

Billing. Monthly fee, based per GB of storage allocated. Allocated storage is measured every 30 minutes, averaged across the month.
#15..
- 873 -
SOLICITATION # CH16012
135
Tier I Block Storage - Local Only (IC-STO-T1A-LOC)
Block storage with a Latency Service Level of 10ms that is maintained in a single data center with no replication capability.

Billing. Monthly fee, based per GB of storage allocated. Allocated storage is measured every 30 minutes, averaged across the month.

Tier II Block Storage - Local Only (IC-STO-T2A-LOC)
Block storage with a Latency Service Level of 20ms that is maintained in a single data center with no replication capability.

Billing. Monthly fee, based per GB of storage allocated. Allocated storage is measured every 30 minutes, averaged across the month.

Tier III Block Storage - Local Only (IC-STO-T3A-LOC)
Block storage with a Latency Service Level Objective of 40ms that is maintained in a single data center with no replication capability.

Billing. Monthly fee, based per GB of storage allocated. Allocated storage is measured every 30 minutes, averaged across the month.

Tier 0 Block Storage - Replicated (IC-STO-TOA-REP)
Data storage with a Latency Service Level of 3ms that is persisted in Customer’s primary Virtustream data center, and replicated to a second data center with an interval to provide a Recovery Point Objective (RPO) of 15 minutes.

Service Level. Recovery Time Objective (RTO) of one (1) hour, and a Recovery Point Objective (RPO) of fifteen (15) minutes.

Billing. Monthly fee, based per GB of storage allocated.

Tier I Block Storage - Replicated (IC-STO-T1A-REP)
Data storage with a Latency Service Level of 10ms that is persisted in Customer’s primary Virtustream data center, and replicated to a second data center with an interval to provide a Recovery Point Objective (RPO) of 15 minutes.

Service Level. Recovery Time Objective (RTO) of one (1) hour, and a Recovery Point Objective (RPO) of fifteen (15) minutes.

Billing. Monthly fee, based per GB of storage allocated.

Tier II Block Storage - Replicated (IC-STO-T2A-REP)
Data storage with a Latency Service Level of 20ms that is persisted in Customer’s primary Virtustream data center, and replicated to a second data center with an interval to provide a Recovery Point Objective (RPO) of 15 minutes.

Service Level. Recovery Time Objective (RTO) of one (1) hour, and a Recovery Point Objective (RPO) of fifteen (15) minutes.

Billing. Monthly fee, based per GB of storage allocated.

Tier III Block Storage - Replicated (IC-STO-T3A-REP)
Data storage with a Latency Service Level Objective of 40ms that is persisted in Customer’s primary Virtustream data center, and replicated to a second data center with an interval to provide a Recovery Point Objective (RPO) of 15 minutes.

Service Level. Recovery Time Objective (RTO) of one (1) hour, and a Recovery Point Objective (RPO) of fifteen (15) minutes.

Billing. Monthly fee, based per GB of storage allocated.

8.12.3 Specify and provide the process to be used for the participating entity to call/contact you for support, who will be providing the support, and describe the basis of availability.

CA APM CA Support can be engaged by telephone or online via CA Support Online at http://www.support.ca.com. CA Support is available 24x7x365. Technical support will be delivered in accordance with customer preference, and will likely include both telephone and email communications. CA Support Online is available 24 hours per #15..
- 874 -
SOLICITATION # CH16012
136
day and allows customers to manage their cases (i.e. log, view, set severity level, update and close cases). CA Support Online also provides various other useful features, for example: a knowledge base search; product and documentation downloads; fixes, service pack and patch downloads; support utilities; automated notifications; access to communities, beta programs; subscriptions for hyper notifications, etc.

While working with CA Support engineers on cases, email communications and a remote access capability can also be used to deliver support services. Secure File Transfer Protocol (FTP) and a secure FTP web client allow for the protected transfer of case file attachments to CA Support, with the web client additionally providing case file transfer protection from CA Support. Files are encrypted and stored securely while on CA systems. Optionally unsecure FTP transfers also are supported." MAA CA Support can be engaged by telephone or online via CA Support Online at http://www.support.ca.com. CA Support is available 24x7x365. Technical support will be delivered in accordance with customer preference, and will likely include both telephone and email communications. CA Support Online is available 24 hours per day and allows customers to manage their cases (i.e. log, view, set severity level, update and close cases). CA Support Online also provides various other useful features, for example: a knowledge base search; product and documentation downloads; fixes, service pack and patch downloads; support utilities; automated notifications; access to communities, beta programs; subscriptions for hyper notifications, etc.

ASM CA Support can be engaged by telephone or online via CA Support Online at http://www.support.ca.com. CA Support is available 24x7x365. Technical support will be delivered in accordance with customer preference, and will likely include both telephone and email communications. CA Support Online is available 24 hours per day and allows customers to manage their cases (i.e. log, view, set severity level, update and close cases). CA Support Online also provides various other useful features, for example: a knowledge base search; product and documentation downloads; fixes, service pack and patch downloads; support utilities; automated notifications; access to communities, beta programs; subscriptions for hyper notifications, etc.

While working with CA Support engineers on cases, email communications and a remote access capability can also be used to deliver support services. Secure File Transfer Protocol (FTP) and a secure FTP web client allow for the protected transfer of case file attachments to CA Support, with the web client additionally providing case file transfer protection from CA Support. Files are encrypted and stored securely while on CA systems. Optionally unsecure FTP transfers also are supported." Google Customers are entitled to direct Google support in additional to whatever Support Services the reseller has to offer. Google has a published Technical Services Support Guide for both Google Apps and Google Cloud Platform that describe the hours of operation, how to get after hours #15..
- 875 -
SOLICITATION # CH16012
137
support, how to set the priority on a case, how to contact Google, and what the target response times are.

AODocs Customers are entitled to direct AODocs support in additional to whatever Support Services the reseller has to offer.

Salesforce Salesforce is proposing the Premier+Success Plan. Your Agency can cover all of its bases with this plan: support, training, and administration. It’s what the + is all about. The Premier+ Success Plan gives you all the benefits of our Premier success plan, including a support rep assigned to your organization, priority case routing, 1-hour response time, 24x7 phone support, unlimited usage of our entire online course library, plus one very helpful addition: access to your own team of expert Salesforce administrators. This lets you focus on design and management while we support your configuration.

The Premier+ Success Plan includes:
ズ Unlimited access to our entire online course catalog of 100+ courses ズ 24x7 toll-free phone support
ズ Priority case queuing and routing
ズ Quick 1-hour response time
ズ An assigned support account rep
ズ Force.com code troubleshooting
ズ Customizable end-user course templates
ズ Premier Success Review to measure usage and trends
ズ Access to our pool of Salesforce Certified Administrators who can configure and maintain your Salesforce edition
ズ Premier Success Review to measure usage and trends
ズ Includes administration services: Get configuration help. Request 100+ routine configuration updates like creating users, reports, workflow, and dashboards. You take online administration training to learn the basics, then you tell us your business requirements. Our team of certified administrators updates your Salesforce system.

Subject to the Government Cloud Premier+ Success Plan outlined above, access to systems and permissions which could permit access to Customer Data inside of the Salesforce Government Cloud storing U.S. government, U.S. government contractors, and FFRDC Customer Data will be restricted to Qualified U.S. Citizens. Qualified US Citizens are individuals who are United States citizens, and are physically located within the United States when accessing the Salesforce Government Cloud systems; and have completed a background check as a condition of their employment with Salesforce.Severity Levels & Response TimesIssues will be categorized and handled according to an assigned severity level, as follows:

Level 1 - Critical
Response time: 1 Hour*
Critical production issue affecting all users, including system unavailability and data integrity issues with no workaround available.

Level 2 - Urgent
Response time: 2 Hours*
Major functionality is impacted or significant performance degradation is experienced. Issue is persistent and affects many users and/or major functionality. No reasonable workaround available.

Also includes time-sensitive requests such as requests for feature activation or a data export.

Level 3 - High
Response time: 4 Hours**
#15..
- 876 -
SOLICITATION # CH16012
138
System performance issue or bug affecting some but not all users. Short-term workaround is available, but not scalable.

Level 4 - Medium
Response time: 8 Hours**
Inquiry regarding a routine technical issue; information requested on application capabilities, navigation, installation or configuration; bug affecting a small number of users. Reasonable workaround available. Resolution required as soon as reasonably practicable. *24/7 Severity 1 and 2 coverage includes weekends and holidays **Severity 3 and 4 target response times include local business hours only and exclude weekends and holidaysOverview of Support Escalation PathAn issue is received by Tier 1 Support Rep who will document details and attempt to resolve. In the event that the problem cannot be resolved at this level, case will be escalated to a Tier 2 Support Rep for further analysis. Occasionally, a Case may be received for which technical understanding or indicated fix exceeds the responsibility of Support. These issues are quickly brought to the attention of R&D through the Tier 3/QA channel.

Throughout the lifecycle of the case, customer will receive regular updates from the case owner at regular intervals based on case severity. Customers can work with their Support Account Specialist (SAS) if assigned through the purchase of Premier Support as a point of escalation should the need arise.

Step-by-step escalation path:
•Case is logged via Help Portal, Phone or through Live Chat •Depending on the functional area of the case a Tier 1 Support Rep from the appropriate skill group responds to the user confirming that the issue has been received and is in process. •Tier 1 analyzes the issue from the perspective of customer provided information in the case. •Tier 1 makes contact (phone or email) with the customer and gathers more information including any access needed to reproduce the problem.
•With all necessary information gathered, Tier 1 will attempt to resolve the issue with the customer. •When resolution cannot be immediately achieved, Tier 1 Support Rep will escalate to a Tier 2 Support Rep in the same skill group. Tier 2 Support Rep will take all steps necessary to reproduce, understand and resolve the issue.
•Technical Issues that cannot be resolved by the Tier 2 Support Rep are escalated to Tier 3. This escalation will provide a deeper layer of analysis with the possibility of escalation to QA/Development for Application issues and Escalation to Operation Services for Infrastructure issues as needed.
•Functional Issues that cannot be resolved by Tier 3 are escalated to Product Management. •If it is an application Issue, necessary Development work will be performed, tested and a patch created.
•Issue Fix Submitted to Operations for Infrastructure issues. •Issue Fix deployed.
•Tier 2 Support Rep confirms success of Bug Fix in addressing the problem. •Tier 2 Support Rep contacts customer and confirms that the issue is Resolved. •Case is closed"
ServiceNow Customer Support is provided by ServiceNow 24 hours a day, 7 days a week, including all holidays. Customer may contact ServiceNow using one of the following means:
•Support Portal at https://hi.service-now.com/. Customer may get login access to this self-service portal by contacting its ServiceNow administrator.
•Phone using one of the numbers at http://www.servicenow.com/support/contact-support.html.

Customer shall contact ServiceNow’s authorized reseller in accordance with its agreement with the reseller.
#15..
- 877 -
SOLICITATION # CH16012
139
See the “Customer Support Policy” contained within the “Subscription Service Guide” included with this response for more information.

Docusign Enterprise Premier support provides 24x7 Live Phone Support.

QTS Dedicated support staff is available 24x7x365 to generate and track incidents to resolution QTS Customer Portal provides round-the clock visibility into your environment VMware Global Support Services
FireEye FireEye offers 24/7 technical support and its technical support centers around the globe in a follow-the-sun model.
•Milpitas, CA, USA (UTC -7 hours)
•Reston, VA, USA (UTC -4 hours)
•Dublin, Ireland (UTC)
•Singapore (UTC +8 hours)
•Sydney, Australia (UTC +10 hours)
FireEye Support is available via telephone, email, live chat, and web.

Phone: U.S: +1 877-347-3393 (877 FIREEYE)
Email: [email protected]
Web Support: https://www.fireeye.com/support/get-support.html Support Programs: https://www.fireeye.com/support/support-programs.html VirtueStream Virtustream will work collaboratively with State of Utah to develop / maintain a relationship structure that leads to quick resolution, regularly scheduled status meetings, and quarterly business reviews.

Virtustream will assign multiple key personnel to State of Utah including an Executive level sponsor. The primary point of contact will be Virtustream’s Technical Account Manager (TAM).

The TAM is responsible for daily activities/interaction and understanding/ensuring State of Utah’s objectives are met at a minimum. Optimizations and other improvements to processes/standard operating procedures are delivered by the TAM in coordination with other Virtustream staff working behind the scenes. The TAM will coordinate day-to-day operation, service level management and SL reporting.

The following outlines Virtustream’s approach to service desk escalation and issue resolution.

Escalations within Virtustream
Virtustream will provide Level 0 and Level I support and first call resolution where possible, as determined by Virtustream. Where first call resolution is not possible, the Virtustream Service Desk provides incident management for Incidents and Urgent Service Requests escalated to Level II and Level III resources as defined below.

In the event that Virtustream’s Response to an Incident is not acceptable to the Customer, Customer can contact the Virtustream Service Desk and request escalation to the head of the Service Desk. Virtustream shall, upon receipt of any such request, immediately escalate the issue to the head of the Service Desk or technical team as appropriate.

Service Request Prioritization
Service Requests are assigned a priority of either ‘Urgent’ or ‘Standard’ and are queued for fulfillment with the corresponding priority. All Service Requests will be reviewed by the Virtustream Service Desk, who will determine the appropriate priority to assign with collaboration of Customer.

Incident Prioritization
All Incidents that are reported to the Virtustream Service Desk, or that Virtustream otherwise becomes aware of, will be initially assigned a priority by the Virtustream Service Desk as set forth below. Internal escalation for Incidents to Level II and Level III resources are based on the priority level assigned to the Incident.
#15..
- 878 -
SOLICITATION # CH16012
140
Incident Prioritization
Priority/
Severity Definition
Response Time Service
Level
1 Major part of the system is unavailable/not operating correctly, affecting multiple users. No workarounds in
place and business operations are not possible.

Or
Incident has a critical impact on the business (e.g.,
loss of the Exchange Production server impacting all
users).
30 minutes
Response time
will be within
indicated time
beginning from
when the
customer
creates a ticket
or a monitoring
event is
validated.

Additional
resources are
engaged via
Virtustream’s
on Call
Process.
2 Part of the system is unavailable/not operating
correctly, affecting users in a single function. No
workarounds in place and business operations in this
function are not possible/severely impacted.

Or
Incident has a serious impact on part of the business
(e.g., a configuration change is impacting a small
subset of users).
60 minutes
3 Part of the system is unavailable/not operating
correctly, affecting users in a single
function. Workarounds in place, but business
operations are impacted, although not severely.

Or
Incident has a temporary impact on users and is non
critical or is a development issue (e.g., email is slow to deliver)
4 hours
4 Incident that is causing inconvenience to the
business, but not impacting operations.

Or
Incident has a minor impact on users or business, or
issue is a request for further information
1 business
day
Virtustream will assign a Technical Account Manager starting on day 1 of the contract. The Cloud Platform, Cloud Cover (managed services), and Cloud Security team’s will also be assigned to State of Utah upon execution of the contract. All of the individuals assigned to the State of Utah account have significant years of experience in their fields.

The TAM assigned to State of Utah will provide reports ad hoc but also on a quarterly basis. The quarterly business reviews include but are not limited to performance measurements, service requests. On a monthly basis, the TAM will provide a consumption report detailing the items consumed by each virtual machine. This delivers a very granular cost breakdown to State of Utah which will allow the data to be carved multiple ways. Also, information can be gathered by the State of Utah team any time they want from the xStream portal. State of Utah can also issue a ticket to Virtustream operations center to request information as well. The TAM will conduct scheduled meetings and be in constant contact with State of Utah. The TAM is an extension of the State of Utah team.

Virtustream defines a Root cause analysis (RCA) as the formal process, documented in writing by Virtustream and approved by Customer, to be used by Virtustream to diagnose problems at the lowest reasonable level which includes a report of the corrective action to be taken and defined timelines for corrective actions, which shall eliminate, to the extent reasonably possible, repeat failures. The following details the RASCI for RCA’s:
#15..
- 879 -
SOLICITATION # CH16012
141
Role/Function Customer Virtustream
Request Root Cause Analysis tickets by contacting the
Virtustream TAM (Customer requests Incident report/Problem record)
R/A S
Document, track and manage all Problem tickets using ITSM system S R/A
Provide Problem management review and Root Cause Analysis (RCA) for all in-scope P 1 Incidents (preliminary report within 48 hours; final within 15 calendar days)
S R/A
Provide Problem management and RCA of identified Problems (e.g., reoccurring events, alerts) - investigate and diagnose S R/A

8.12.4 Describe the consequences/SLA remedies if the Respondent fails to meet incident response time and incident fix time.

CA APM In the event that the Service Level Availability committed decreases below the Threshold for Service Availability Default, Minor or Major, Customer may be entitled to take action as outlined in the contract.

MAA In the event that the Service Level Availability committed decreases below the Threshold for Service Availability Default, Minor or Major, Customer may be entitled to take action as outlined in the SaaS Listing.

CA Agile In the event that the Service Level Availability committed decreases below the Threshold for Service Availability Default, Minor or Major, Customer may be entitled to take action as outlined in the contract.

ASM In the event that the Service Level Availability committed decreases below the Threshold for Service Availability Default, Minor or Major, Customer may be entitled to take action as outlined in the SaaS Listing.

AODocs If AODocs does not meet the SLAs, the customer will be eligible to receive a service credit.

Virtru Email [email protected]
ServiceNow ServiceNow or its authorized reseller, as applicable, will use reasonable efforts to meet the target response times and target level of effort stated in the “Customer Support Policy” contained within the “Subscription Service Guide” included with this response for more information.

SAP Ariba As part of the Cloud subscription, Ariba provides Customer Support services to help diagnose, troubleshoot and resolve functional and technical problems for all users. Ariba’s award winning customer service staff can be contacted via phone (toll free), email or Web form Monday-Friday, 24/5 in multiple time zones and languages. Ariba Connect, our online support portal provides 24/7 support for service and enhancement requests. Customers may also choose to purchase ExpertCare. ExpertCare provides each customer with a named contact within Ariba that acts as an ExpertCare Manager for the customer and is knowledgeable about customer specific business process, configurations and customizations.

Fieldglass Our best practice is that the program office (managed by the State of Utah or your chosen MSP) provides first-level support for end users including hiring managers, approvers, suppliers, etc.

Fieldglass' multi-lingual helpdesk is available by phone, email or web form 24x7x365. Our English language helpdesk is located at our corporate headquarters in Chicago. Merlin Information Systems Ltd., a UK entity, provides helpdesk service in French, German and Japanese from its Manila, #15..
- 880 -
SOLICITATION # CH16012
142
Philippines and Debrecen, Hungary locations. We are prepared to assist with technical, functional or administrative questions about the Fieldglass application.

Hanna Customer key users have several options to communicate incidents into SAP Managed Cloud Delivery.

Primary entry point:

The best and easiest way to get an incident resolved is by creating and sending incidents directly to SAP Managed Cloud Delivery Service Teams via the SAP Support Portal. This communication channel ensures that all support teams are notified and therefore is the most preferred way to contact SAP on support/service related issues.

Alternative option to the primary entry point:

Customer key user can also communicate incidents by telephone via the SAP Customer Interaction Center (CIC) (e.g. in case of facing challenges in submitting an incident via the SAP Support Portal).

Exceptional option:

During business hours, customer key user may also communicate urgent incidents by contacting designated SAP Managed Cloud Delivery service contacts like Customer Engagement Service Manager (CESM) to create an incident on behalf of the customer. However – as already mentioned - this is not the preferred way of communication, but only should be applied in exceptional cases.

Support times are contractually agreed upon with each individual customer organization.

The SAP Managed Cloud Delivery (MCD) Service Desk and the Service Teams are spread across Europe, Americas and Asia.As a result of the global set-up across several time zones, SAP MCD Service Desk is able to provide worldwide 24/7 support to customers, throughout the entire year and according to the “Follow-The-Sun-Principle”.

Hybris Key service levels are delivered through service level objectives, which provide for uptime of 99.9 for the hosted environment. Furthermore, monitoring is in place to report and alert upon uptime and response times of the environment to ensure best possible performance.

The following support is specifically for the SAP Hybris Commerce, Cloud Edition.

Customers can contact SAP Hybris Commerce, Cloud Edition technical support by toll-free telephone number or e-mail 24 hours per day, seven days per week. Based on the priority level, SAP Hybris’s support personnel will attempt to provide Customer with remote assistance for reported issues.

Contact information for technical support is as follows:

Toll-Free Telephone Number: 888-944-2664 ext 6
In the event that Customer contacts SAP Hybris support, SAP Hybris shall respond to such reports as follows:

Priority Level :

Characteristics
Response Time
Resolution Time Objective
#15..
- 881 -
SOLICITATION # CH16012
143
Priority Level 1
Fatal Issue: Problem causing the Customer Website to cease from operating.

This situation is generally described as a total failure. 25 Minutes
(7x24)
2 hours
Priority Level 2
Major Issue: Problem causing the Customer Website to experience major problems with it ability to operate. This situation exists when the Customer Website is partially failing however it is still able to function. 25 minutes
(7x24)
12 hours
Priority Level 3
Degradation of Performance: Problem affecting only certain non-critical functions of the Customer Website. This situation is occurring when the Customer Website is usable but has certain limited functions. 2 hours
1 business day
Priority Level 4
Minor Issue: This situation includes all other non-critical problems. It is present when the Customer Website is usable however the problem results in a minor issue affecting it.
2 hours
3 business days
Upon receipt of a service call classifying the priority level, SAP Hybris will deploy commercially reasonable efforts to respond and assist in the resolution of the problem.

Application support for code developed by Customer or its implementation partner will be provided by Customer or Customer’s implementation partner.

If SAP hybris provides SAP Hybris Commerce, Cloud Edition support to a Customer as requested by Customer, and it is subsequently revealed that SAP Hybris was not responsible for the problem or the problem was not classified correctly by the customer, Customer will pay SAP Hybris on a time and materials basis for such support and assistance.

SuccessFactors
VMware Our Availability Commitment for VMware AirWatch is 99.9%. The SLAs are subject to the terms of the applicable Terms of Service for VMware AirWatch.

If the Availability of a class of service that you purchase is less than the associated Availability Commitment, then you may request Service Credits for that affected class of service. Availability in a given month is calculated according to the following formula:
“Availability” = ((total minutes in a calendar month – total minutes Unavailable) / total minutes in a calendar month) x 100
A class of service will be considered “Unavailable,” subject to the Service Level Agreement Limitations set forth below, if VMware’s monitoring tools determine one of the following events has occurred (“SLA Event”). The total minutes that a class of service is Unavailable for a particular SLA Event is measured from the time that VMware validates the SLA Event has occurred, as defined below, until the time that VMware resolves the SLA Event such that the Service Offering is Available to you. If two or more SLA Event occurs simultaneously, the SLA Event with the longest duration will be used to determine the total minutes Unavailable. #15..
- 882 -
SOLICITATION # CH16012
144

1. Each of the following will be considered an SLA Event for the Dedicated Cloud, Virtual Private Cloud, or Virtual Private Cloud OnDemand Services:
a) Any of the network interfaces of the Service Offering Network are unavailable for more than three (3) consecutive minutes. The “Service Offering Network” means the network that extends from the network interfaces of physical host servers for a class of service to the outside network interfaces providing the Service Offering's public internet connectivity. b) The data store(s) associated with your block level storage for a class of service are unavailable for more than three (3) consecutive minutes.
c) The self-service console, available at https://vchs.vmware.com, cannot successfully authenticate a simulated user for more than five (5) consecutive minutes. d) Your running virtual machines for a class of service become inaccessible for more than five (5) consecutive minutes due to physical host server failures.

2. Each of the following will be considered an SLA Event for the Data Protection Service:
a) The backup storage repository associated with your Data Protection service is unavailable for more than three (3) consecutive minutes.
b) The start time associated with your backup scheduling window is missed for longer than thirty (30) consecutive minutes.
c) The standard 24-hour recovery point objective (RPO) is missed for any virtual machines actively enrolled in the service.
d) Any restore operation for a virtual machine fails to complete due to backup infrastructure failures.

3. Each of the following will be considered an SLA Event for the Disaster Recovery Service:
a) Any of the network interfaces of the Service Offering Network are unavailable for more than three (3) consecutive minutes.
b) The data store(s) associated with your block level storage for replication is unavailable for more than three (3) consecutive minutes.
c) The self-service console, available at https://vchs.vmware.com, cannot successfully authenticate a simulated user for more than five (5) consecutive minutes. d) Your failed-over virtual machines for a class of service become inaccessible for more than five (5) consecutive minutes due to physical host server failures. e) Any built-in service functions for failover testing, planned migration, or live failover and recovery result in virtual machine replicas not powering on in less than 4 consecutive hours from the time a request is acknowledged and approved by VMware.

4. The following will be considered an SLA Event for the Object Storage service:
a) A more than five (5) percent Error Rate for more than ten (10) consecutive minutes – where “Error Rate” means the number of valid requests that result in a response with HTTP Status 500 and Code “Internal Error” divided by the total number of valid requests during each five-minute period.

FireEye FireEye shall use commercially reasonable efforts to correct any reproducible programming error in the Software attributable to FireEye, employing a level of effort commensurate with the severity of the error, provided, however, that FireEye shall have no obligation to correct all errors in the Software.

8.12.5 Describe the firm’s procedures and schedules for any planned downtime.

CA APM Planned downtime is scheduled, and customers are notified MAA Customers are notified at least two weeks in advance for planned downtime.

CA Agile Regularly scheduled maintenance (planned downtime for upgrades and maintenance) where the customer has been given at least eight (8) hours notice does not count as downtime. Unscheduled maintenance, in which the Rally Service is unavailable and #15..
- 883 -
SOLICITATION # CH16012
145
advance notice was not provided to customers, will be counted against the up-time SLA.

ASM Customers are notified at least two weeks in advance for planned downtime.

Google Google's maintenance routines are done in a manner that ensures there is no need to schedule any downtime.

AODocs AODocs's maintenance routines are done in a manner that ensures there is no need to schedule any downtime.

Salesforce There are two types of maintenance at Salesforce: System Maintenance and Release Maintenance. System Maintenance is for sustaining the performance, reliability, security, and stability of the infrastructure supporting our services. When maintenance is scheduled, the specific timing and availability to be expected during that time will be communicated to all customers approximately one week in advance of the maintenance window upon login to Salesforce and via a posting to http://trust.salesforce.com/trust/maintenance. In the event of planned maintenance where the services will be unavailable for more than four hours, or that requires customer action in advance, Salesforce endeavors to communicate via email several months in advance. Please Note: If emergency system maintenance is required, customers may be notified less than one (1) week in advance.

Major Release Maintenance is for upgrading the services to the latest product version to deliver enhanced features and functionality. Major release dates and times are posted on http://trust.salesforce.com/trust/maintenance approximately one month before release to Sandbox instances. An email notification and blog post regarding Sandbox preview instructions is also sent approximately one month prior to upgrading Sandbox instances. Email notification of major release dates is sent one month prior to upgrading non-Sandbox instances. The Release Notes document describing the new features and functionality is posted in Help & Training one month prior to upgrading non-Sandbox instances. Final release reminders are communicated to all customers approximately one week prior via email and upon logging into Salesforce.

Major release maintenance occurs three times per year. The instance will be unavailable for up to five minutes during the release window.

Patch Releases and Emergency Releases are used to deliver scheduled and ad hoc application fixes. Patch releases are scheduled weekly and are usually deployed to instances on Tuesday, Wednesday or Thursday, with release to Asia-Pacific instances the following day. Emergency releases are conducted on an as-needed basis and can occur any day of the week. Whenever possible, patches and emergency releases are deployed during off-peak hours and without downtime.

ServiceNow ServiceNow notifies customers at least 10 days prior to scheduled maintenance for infrastructure network, hardware or software that might impact service. ServiceNow targets no more than two hours of downtime due to scheduled maintenance per month.

Docusign Continuous availability is a necessity for mission-critical business applications in a global world.

It’s also a key component of the xDTM standard. Zero maintenance downtime means that DocuSign is always available, regardless of the time-of-year, day-of-week, time zone or country.

DocuSign’s Carrier Grade System Architecture, a first in SaaS, eliminates maintenance downtime, and ensures the highest level of performance resiliency and data integrity. This is accomplished through a combination of advanced high availability architecture, native DocuSign engineering, and both specialized and commodity hardware and software components.

SAP Ariba Scheduled downtime windows for maintenance, upgrades or new releases are provided in the Service Level Agreement. Our uptime percentage takes scheduled maintenance into account.
#15..
- 884 -
SOLICITATION # CH16012
146
Fieldglass Fieldglass' maintenance windows are as follows:
• Friday maintenance window of 11:00 p.m. – 2:00 a.m. Central (Daylight Savings Observed)
• Extended maintenance windows are taken with five days advance notice and are typically as follows:
- Second Friday of each of the following months: March, July, and November - Critical service packs, security issues, and other emergencies will still be taken if needed with communication as reasonable
Hanna SAP performs periodic scheduled maintenance activities to maintain security patching levels, application transports, fixes, network maintenance, and other scheduled proactive activities during scheduled maintenance window (agreed to by the customer). The predefined maintenance window is 4 hours per month (for example first friday of every month)
SAP may perform additional maintenance services on the Customer’s Computing Environment in agreement with Customer, with at least ten (10) Business Days prior notice (which notice may be by email to Customer’s nominated contact or posting) of the date and anticipated length of the maintenance window, and provided that SAP adheres to Customer’s reasonable requirements as to the timing and duration of any such maintenance.

SAP reserves the right to apply critical application and operating system security patches in compliance with the SAP Global Security Team’s reasonable recommendations. SAP will use reasonable endeavours to provide Customer with forty-eight (48) hours advance notice regarding the critical patch deployment unless a shorter notice period is required in order to address a critical security issue. Non critical security issues shall be dealt with by Scheduled Maintenance
Hybris Typically they are two-hour maintenance windows, with no downtime or on occasion maintenance periods require between 2 to 10 minutes of downtime.

Other rare periods which may require further maintenance could have downtime between 30 min to 4 hours.

SuccessFactors To address issues such as security patches, hot fixes, updates, equipment upgrades, etc., we has established maintenance windows to perform scheduled maintenance. Scheduled maintenance events typically occur during only one of these windows, roughly every 6-8 weeks. A 14-hour/week window will be used for all critical / most emergency patches and all bug fixes (will include tech stack and application patch sets).

We define scheduled downtime maintenance windows as:

Midnight to 7 a.m. Fri-Sat local time + midnight to 7 a.m. Sat-Sun local time (dependent upon the data center where the customer data is hosted) We will provide customers with at least a 2 days’ notice if a larger maintenance window is needed. Typically, for HCM / LMS products, customers receive a 3-5 day notification (via popup on login). When the system is not available during the maintenance there is a notification page displayed for the outage timing.

We plan maintenance events conservatively, and use them sparingly.

VMware VMware IaaS Services
Customers are notified in advance of any expected downtime with vCloud Air services. In the event of an unexpected outages or disruption to the service, customers will be notified via their preferred contact means as soon as possible as per the terms outlined in the Service Level Agreement.

VMWare AirWatch
#15..
- 885 -
SOLICITATION # CH16012
147
We maintain standard SLAs with our customers to provide for downtime • SLAs are defined within the AirWatch Hosted Services Policy Per our Hosted Services Policy, we provide outage credits in the event of a service interruption:
• ""AirWatch warrants it will provide Availability of the Hosted Services 99.9% of the calendar month (the “Service Level”). To the extent that the Hosted Services fail to conform to the Service Level, Customer may request service credits (“Outage Credits”) as provided herein. A failure or lack of Availability for any period of time of at least one minute during which Customer is unable to utilize the Hosted Services due to AirWatch’s failure to provide Customer with the specified services constitutes an “Outage”. All Outage measurements will be rounded up or down to the nearest one minute increment, with increments equal to or greater than 30 seconds being rounded up to the next minute. Outage Credits are based on cumulative periods of Outage over a calendar month. Final determinations of the length of the cumulative periods of Outage over a calendar month shall be based on AirWatch’s internal monitoring equipment and records. Outage Credits will be taken against only the Hosted Service fees for the month in which the Outage occurred."" FireEye FireEye undertakes reasonable efforts to ensure that the service availability maintains a minimum of 99.9% availability. However there are circumstances where the system will be down due to the following:
a) “Scheduled Maintenance Period" is the period during which weekly scheduled maintenance may be performed, or a maintenance window otherwise mutually agreed upon by FireEye and Customer. Customers receive notification and announcements for any scheduled maintenance period.
b)"Emergency Maintenance" means any time outside of Scheduled Maintenance that FireEye is required to apply critical patches or fixes or undertake other urgent maintenance. If Emergency Maintenance is required, FireEye will contact Customer and provide the expected time frame of the Emergency Maintenance.

VirtueStream Virtustream planned downtimes start with the weekly change board meeting by the Platform Operations group. All requested changes whether requested by customers or by the operations staff are reviewed, if approved then scheduled. Depending on the nature of the change, routine and planned, the affected customers are notified as early as possible and adjusted based on customer needs. For example, if the scheduled downtime falls on a weekend that a customer plans to do their quarterly close, Virtustream will reschedule. An emergency/urgent change may have less notice for the customers. But in all cases, customers will be notified of the upcoming downtime multiple times, start of the downtime will be specified and the expected duration. Once the downtime is complete, another notification will go out to the affected

8.12.6 Describe the consequences/SLA remedies if disaster recovery metrics are not met.

MAA Defined monetary penalties are provided in the SaaS Listing document CA Agile See 8.12.4
ASM Defined, monetary penalties are provided in the SaaS Listing document Google Customers will be eligible to receive a service credit.

AODocs Customers will be eligible to receive a service credit.

Salesforce Customer data, up to the last committed transaction, is replicated to disk in near-real time at the designated disaster recovery data center, and backed up at the primary data center. Backups are performed daily at primary data center facility without stopping access to the application. #15..
- 886 -
SOLICITATION # CH16012
148
For business continuity purposes, Salesforce supports disaster recovery with a dedicated team and a 4 hour recovery point objective (RPO) and 12 hour recovery time objective (RTO).

Additional details can be provided with the execution of an NDA between Salesforce and your Agency.

In terms of remedies, the Salesforce Service Terms incorporated as part of the End User Licensing Agreement indicate the following regarding Termination allowing customers to terminate the contract at time of renewal:
""User subscriptions will automatically renew for additional periods of one (1) year at the list price in effect at the time of renewal unless You give Your Reseller notice of termination at least 30 days prior to the end of the relevant subscription term.""
ServiceNow See 8.12.2 for the Availability SLA.

Docusign See 8.12.4.

SAP Ariba Please see the details provided in 8.12.4 response.

Fieldglass Please see the answer to Question 8.12.4, above.

Hanna Penalties and remediation for SLA violations are detailed in the Statement of Work.

Hybris Our SLA provides tenant remuneration for losses they may incur due to outages within the infrastructure.

VMware For VMware there is a service credit reimbursement structure outlined above. We have detailed recovery procedures that include defined steps, roles and responsibilities and accountable personnel to help ensure streamlined disaster recovery.

As conversations progress, AirWatch can provide specific details regarding our disaster recovery timelines and strategies.

FireEye FireEye has established formal Business Continuity and Disaster Recovery plan, as well as Incident Response and Recovery Plans to help maximize availability of critical customer-impacting services.

8.12.7 Provide a sample of performance reports and specify if they are available over the Web and if they are real-time statistics or batch statistics.

CA APM A report of the Service’s measured monthly SLA is available to Customer upon request
MAA Reports are generated periodically and ad hoc. Customers can use their support.ca.com credentials to access private trust site which contains SLA details for the subscribed service.

CA Agile Realtime and historical system status can be viewed at https://status.rallydev.com ASM Reports are generated periodically and ad hoc.

Google Google does not produce customer specific performance reports. The web based Apps Status Dashboard displays real time System Availability Details. Customers can works with their Google reseller to design a reporting system using Google Sheets based on which incident actually impacts their end users. Service availability issues are isolated to specific applications and specific geographically locations and thus would not impact all of your users or your account at all.

AODocs AODocs does not produce customer specific performance reports. Customers can visit the AODocs status page: status.aodocs.com. AODocs provides a set a usage reports and AODocs team can also work with the customer to build specific reports Virtru We have realtime monitoring of both server side operations and client side interactions.

Salesforce System Performance ReportsOur track record speaks for itself—Salesforce has maintained high levels of availability across all Salesforce instances since inception. As an on-demand vendor #15..
- 887 -
SOLICITATION # CH16012
149
providing daily service-quality data on a public website (http://trust.salesforce.com), Salesforce proves that it is a leader in availability. And by making its track record completely transparent, salesforce.com proves it is worthy of customers’ trust.

A few metrics that we maintain (and publish publicly via our trust.salesforce.com website) are:
-Daily Transaction Counts
-Daily Transaction Speeds
Salesforce is hitting almost 4 billion transactions a day. These transactions are a mix of user interacting with the system and API calls where other systems are bi-directionally interacting with Salesforce.

Our Average Transaction Time hovers between 150 and low 200 milliseconds.

Some additional published metrics are uptime & availability, planned maintenance windows, and any performance degradation.

The Purchasing Entity can view all of this information and more by going to trust.salesforce.com.

Customers can also log a case to request system uptime reports for the last 6 months and last 12 months, however, system reports are not automatically distributed.

ServiceNow Administrators can view a wide range of performance metrics for their instance and for the machine on which the instance is running, displayed in a graphical format.

Add these graphs and their controls to a home page to monitor the performance of instances.

Some of these graphs are intended for use by ServiceNow Technical Support to troubleshoot performance issues or help tune your system for maximum efficiency. Each graph enables admins to filter the data by using different measurements, such as maximum and minimum values, means, and medians. The available graphs reflect performance in eight functional areas of ServiceNow. •Database
•Discovery
•Disk Partitions
•Linux Stats
•Logging
•MySQL Overview
•Node Metrics
•Replication
•ServiceNow Servlet
Docusign For real-time data, please visit the DocuSign Trust site (trust.docusign.com) for additional information.

SAP Ariba We provide a monthly report to customers describing the system availability percentage. Depending on the cloud service, these reports are made available either by email, through the cloud service or through an online portal.

Fieldglass Fieldglass executes comprehensive performance testing prior to each release to production. Performance testing is executed in accordance with the Fieldglass Performance Testing Standards document. Types of testing performed include regression, load, stress, and endurance. Results are reviewed and approved by senior members of development, QA, IT, and the Chief Architect. Our standards document can be shared upon request while under NDA.

During peak periods, Fieldglass will see 12,000 concurrent users with page response times averaging 400-700ms. Across all of 2015, we have a 99.4% success rate in meeting or exceeding all monthly customer SLAs for page response times.

Hanna Performance reports on HEC are available to the customer via the SAP One Portal page. It provides the customer among other things: operational info about systems being managed, incident management, security management, user specific settings, planned downtime, etc.
#15..
- 888 -
SOLICITATION # CH16012
150
SuccessFactors Customers can view hosting performance and availability information real-time at the following link: https://support.successfactors.com/Service_Status.

VMware VMware IaaS Services
VMware vCloud Air provides Web-based capacity reporting via the MyVMware portal. For detailed performance reporting, participating entities may wish to acquire VMware vRealize Operations suite, which is a non-Web based software product which provides real-time and historic performance statistics for vCloud Air.

VMWare AirWatch
We can provide sample performance testing documents that validate our ability to handle anticipated workload under real-world conditions as conversations progress.

We have a dedicated team for scalability and performance testing. They work with both internal teams and customers to conduct testing.

FireEye This is not applicable as each of the FireEye offerings are classified as SaaS. The performance of the FireEye cloud environments are continuously monitored and managed by the cloud service operations (CSO) team to ensure secure operation and availability.

VirtueStream Service Levels will be calculated and measured monthly by Virtustream on a calendar month basis and reported each month for the previous month. The reports will be provided to Customer by the tenth (10th) working day of the month following that to which such report relates, commencing on the second (2nd) month following the Service Start Date and each month thereafter. The monthly service level report will contain at least the following items: (i) Uptime statistics for the month concerned; (ii) an analysis of reported incidents over the previous month, broken down by type for discussion; (iii) action plans for items giving rise to concern; (iv) comments and observations on any issues arising from Virtustream’s performance monitoring activities; (v) recommendations on service delivery strategies to maintain or enhance the service level; and (vi) review of general business requirements (“Service Level Report”). Cloud Platform Services (CPS) has its own specific service levels as described in this document. Cloud Cover Services (CCS) has service levels that pertain to the CCS offerings and are reported separately. Not all Virtustream customers have CCS but all Virtustream customers use CPS.

8.12.8 Ability to print historical, statistical, and usage reports locally.

CA APM A report of the Service’s measured monthly SLA is available to Customer upon request
MAA Same as above.

CA Agile This is available upon request.

ASM Customers are sent monthly SLA reports.

Google See 8.12.7
AODocs See 8.12.7
Virtru We may be able to publish reports locally for customers.

Salesforce Trust.salesforce.com is the Salesforce community’s home for real-time information on system performance and security. On this site you'll find:
ズ Live and historical data on system performance
ズ Up-to-the minute information on planned maintenance
ズ Phishing, malicious software, and social engineering threats ズ Best security practices for your organization
ズ Information on how we safeguard your data
#15..
- 889 -
SOLICITATION # CH16012
151
In addition to the Salesforce Trust site (http://trust.salesforce.com/trust/status) to monitor uptime and performance, your Agency will also have access to a System Overview, which will help you monitor performance and usage of your own Salesforce org. This overview includes:

Schema - # and % of custom objects and data storage
Business Logic - # and % of Rules, Apex triggers and classes, as well as % of code used License Usage
API Usage - # and % of requests in the last 24 hours
User Interface - # and % of custom apps, sites, flows, custom tabs and pages Portal Usage
The above list is of all the possible metrics that you may have in your system overview.

ServiceNow Yes. Instance performance can be viewed locally.

Customers can use the Customer Support Portal to obtain information about the real availability of all their instances. Real availability is the percentage of production time that an instance is up and available for use.

Docusign Yes. DocuSign offers a number of pre-defined reports to provide detail and summary information on DocuSign usage in the account, such as by Envelope, Recipients, Velocity of signing, and User Access among many other reports. Users can modify report criteria such as data ranges and reports can be scheduled to execute. All reports can be downloaded or Printed locally.

QTS Yes
SAP Ariba User access is browser-based, so printing is enabled via the printers that are setup on the user's workstation. There are several “Print” buttons throughout the applications that will open an additional browser window with a printerfriendly format of the screen.

Fieldglass Any user with the correct permissions in the system, as specified by the State of Utah, may locally print historical, statistical, and usage reports.

Hanna Yes, available
VMware VMware IaaS Services
At the present time, VMware does not provide end user customers with specific reports for SLA performance. VMware tracks the overall health of the environment and will report to customers in the event of an outage. However, specific SLA performance is not delivered as a report, but as an alert. It is currently the customer’s responsibility to track downtime within their environments and report to VMware when they believe SLA has been violated.

VMWare AirWatch
Customers can export and print solution data using interactive dashboards (CSV), reports (PDF, XLS, and CSV), the AirWatch Hub (PDF), and event log (CSV) FireEye Each of the offerings in this response have reporting capabilities local to the system and users with the appropriate access rights.

8.12.9 Offeror must describe whether or not its on-demand deployment is supported 24x365.

CA N/A. CA only offers SaaS solutions and manages all aspects of the service. Clients do not have direct access to the environment.

Google Purchasing entities will subscribe to an initial set of user licenses and can provision new user accounts at their convenience within that range of licenses.

Salesforce The Salesforce service can be deployed rapidly since our customers do not have to spend time procuring, installing or maintaining the servers, storage, networking equipment, security products, #15..
- 890 -
SOLICITATION # CH16012
152
or other infrastructure hardware and software necessary. The service is always available and the customer can deploy 24x365 at their own pace.

ServiceNow ServiceNow is supported 24x365.

FireEye The FireEye offerings in this response are SaaS service models. Once the application or service is set up there are no further deployment steps necessary. On-demand Self-service options referenced in section 8.1.2.1 are available 24x365.

VirtueStream Virtustream’s self-service portal allows authorized users to deploy resources on-demand 24x365.

8.12.10 Offeror must describe its scale-up and scale-down, and whether it is available 24x365.

CA N/A. CA only offers SaaS solutions and manages all aspects of the service; monitoring and capacity planning is included as part of the service.

Google Purchasing entities will subscribe to an initial set of user licenses and can provision new user accounts at their convenience within that range of licenses. If the purchasing entity has exceeded or is about to exceed their user count they can place an order for additional licenses at any time.

Scale-down procedures will be handled on the renewal anniversary.

Salesforce The Salesforce cloud based architecture will allow the State to deploy Salesforce solutions rapidly and scale at will for future needs. Customer’s may scale up and add licenses at any point during their annual contract cycle. User counts are examined on an annual basis at time of renewal to ensure the customer had adequate license coverage.

ServiceNow ServiceNow can automatically scale its application servers horizontally by adding or subtracting them to the load balancer pools for a particular instance. Scaling is managed by the ServiceNow Cloud Operations team.

FireEye The FireEye offerings in this response are SaaS service models. FireEye cloud environments are continuously monitored by cloud service operations (CSO) personnel to help ensure optimal resources are allocated to the system 24x365.

VirtueStream Virtustream’s self-service portal allows authorized users to deploy or de-commission resources ondemand 24x365.

8.13 Cloud Security Alliance
Describe your level of disclosure with CSA Star Registry for each Solution offered. a. Completion of a CSA STAR Self-Assessment, as described in Section 5.5.3. b. Completion of Exhibits 1 and 2 to Attachment B.
c. Completion of a CSA STAR Attestation, Certification, or Assessment. d. Completion CSA STAR Continuous Monitoring.

Carahsoft is proposing a number of Service Providers that are members of the Cloud Security Alliance. For example Salesforce is a corporate member of the Cloud Security Alliance (CSA) and contributes to their research and initiatives in a variety of ways. Salesforce has completed the CSA Consensus Assessments Initiative questionnaire (CAIQ), which maps to the CSA Cloud Controls Matrix (CCM) for the Salesforce Services, including Force.com, Analytics Cloud, Communities, Chatter, Sales Cloud and Service Cloud. A copy of the current completed CSA CAIQ can be provided to prospects or customers under NDA upon request.

8.14 Service Provisioning

8.14. 1 Describe in detail how your firm processes emergency or rush services implementation requests by a Purchasing Entity.
#15..
- 891 -
SOLICITATION # CH16012
153
CA APM CA SaaS Ops has a rigorous service introduction and update process that is governed by a central body to ensure adherence. Every step in the process is recorded, tracked and approved.

MAA Standard lead times for requests can be waived for requests that have a critical business need
CA Agile Standard lead times for requests can be waived for requests that have a critical business need such as a data recovery effort
ASM Standard lead times for requests can be waived for requests that have a critical business need such as a data recovery effort
Google Order processing from the time a valid Purchase Order is submitted has a normal turn around time of three business days. Google's sales team can be contacted to expedite orders when needed.

AODocs Order processing from the time a valid Purchase Order is submitted has a normal turn around time of two business days. AODocs's sales team can be contacted to expedite orders when needed.

Virtru Requests are triaged by the support and if an emergency is identified, the request is escalated by the VP of Product who immediatlye assigns the issue as a top priority to a developer. Once completed a hotfix is pushed to production.

Salesforce Subscription-based, On-Demand Service
Salesforce offers on-demand Platform as a Service (PaaS) and Software as a Service (SaaS).

The Salesforce PaaS and SaaS offerings are subscription based and in a per user/month or user/year format billed annually with some of our products offered as total logins per month or by defined number of members billed annually.

Multi-tenancy gives applications elasticity. Force.com applications can automatically scale from one to tens of thousands of users. Processing more than three billion transactions each day, Force.com is used for large-scale deployments. Any application that runs on Force.com is automatically architected to seamlessly scale from 1 user to 100,000 users without the customer having to do anything differently.

All applications (includes mobile, offline and read-only options) and data running on Force.com are deployed to and replicated across multiple data centers in different geographies. Every application, no matter how large or small, gets the full benefits of the backup, failover, disaster recovery, and other infrastructure services required for an organization’s mission-critical applications.

ServiceNow ServiceNow professional services organization responds to customer needs based on each request. Once a purchase order is received from the Purchasing Entity, ServiceNow contacts the Customer to discuss schedules. Resources are assigned based on the project need. If the customer has a priority issue with the system, they would open a ticket and the ticket SLA process would proceed until the issue is resolved. Every ServiceNow customer is supported based on their priority and need.

SAP Hanna Rush or emergency infra provisioning can be request in exceptional situations and would be addressed on best effort priorty basis.

VMware VMware IaaS Services
Initial Order
- Expedite the purchasing process thro
- In the event that an emergency or rush service implementation request is received, VMware’s NASPO program manager will coordinate an expedited response depending on the request. - Expedited Procurement - The Program Manager will directly interface with each company in the supply chain to ensure that the order is processed immediately upon receipt by each party. - Expedited In
Increased capacity
#15..
- 892 -
SOLICITATION # CH16012
154
VMWare AirWatch
Participating Agencies will work with thier dedicated Account Executive to procure addition services for thier AirWatch implementations. Customers can implement and update configurations through the AirWatch administrative console. Additionally, customers can submit ASK tickets for any solution issues that will be tracked through completion.

FireEye FireEye will make every reasonable effort to accommodate rush provisioning requests. The purchasing entity should contact their sales account team who will interface directly with the corresponding implementation team to coordinate an expedited implementation.

8.14.2 Describe in detail the standard lead-time for provisioning your Solutions.

CA APM 48 hour turnaround is typically required
MAA Services do not generally require client action for provisioning. General service catalog requests have a standard 48 hour lead time
CA Agile Services do not generally require client action for provisioning. General service catalog requests, such as a data refresh, have a standard 48 hour lead time ASM Services do not generally require client action for provisioning. General service requests have a standard 48 hour lead time
Google Standard lead time is three business days.

AODocs Standard lead time is two business days.

Virtru The Lead-time for provisioning is no more than 15 minutes ServiceNow ServiceNow provisions new instances or newly purchased users or applications to the instance within 48 hours after receipt of order. If users or applications are required more quickly, the customer controls access to all ServiceNOw capabilities at any time and the true-up occurs thereafter.

Docusign DocuSign is intuitive and easy-to-use. Getting started with our cloud solution, standard console, can be implemented in a day. Adding integrations or API's will add to the complexity of the implementation, but our Professional Services are there to guide you every step of the way.

VMware VMware vCloud Air Services
VMware vCloud Air OnDemand services can be purchased and provisioned in minutes via the vCloud Air self service portal. VMware vCloud Air services other than OnDemand are purchased via purchase order. Purchase orders are typically processed on the day of order. Once services have been purchased the user can immediately begin provisioning services from the self service portal. These services are available within minutes of launch.

VMware vCloud Government Services
VMware vCloud Government Services are purchased via purchase order. Purchase orders are typically processed on the day of order. New environments require an RSA token that is overnighted for 2 factor authentication. Once the user has their RSA token they can begin to provision IaaS Services.

VMWare AirWatch
AirWatch is committed to implementing the solution in a timely manner and we work closely with customers to meet internal deadlines.

Delivery and deployment models will vary depending the hosting model selected by the customer:
• SaaS deployments can be implemented within several days and some within a few hours • Large enterprise deployments with numerous integration endpoints may require additional time Once an order is placed, we immediately begin working with the customer to finalize terms and conditions in a mutually agreeable manner and begin the implementation process.

We schedule regular check-ins for project status updates and to discuss any questions or concerns.
#15..
- 893 -
SOLICITATION # CH16012
155
FireEye FireEye anticipates that the standard lead-time is expected to take no more than five (5) business days. However, FireEye is willing to accommodate an urgent request from NASPO in order to provide adequate service to the participating states.

VirtueStream Virtustream’s standard lead-time for provisioning can vary based of the type of deployment and requirement. Resources can be deployed on-demand basis by State of Utah. For the initial services, there is a standard which is maintained as best practice.

At project kick-off, Virtustream and State of Utah will develop a detailed Implementation Plan, on a system-by-system, week-by-week basis, to ensure that both sides of the relationship have clear expectations on what is to be accomplished, by whom, and in what timeframe. Upon completion of the Implementation Plan, Virtustream and State of Utah will establish a Project Governance Plan and the Project Governance Plan will manage Implementation Plan and ensure alignment at all tiers of the organization, define expectations for status update frequency and formats, and establish an escalation path for issue identification and resolution.

Virtustream offers a dedicated onboarding project team that is charged with migrating clients into the xStreamÆ cloud. This chargeable service includes resources from our platform and SAP teams and development of an onboarding project plan with milestones and deliverables.

Virtustream uses a mature methodology to move clients onto our cloud platform, a process which is continuously refined and improved. A link is established to the client site using a Virtual Private Network (VPN) or Multiprotocol Label Switching (MPLS) circuit, for customer access and testing of migrated systems. Prerequisites are put into place (domain controllers, backup systems, infrastructure, networking, etc.) prior to systems being onboarded, in a defined sequence.

Virtustream SAP basis team members test the systems before turning them over to Ascend for validation. Once on the xStreamÆ platform, Virtustream monitors, backups and maintains the systems. As a component of the Virtustream platform, all core units are replicated to an alternate data center to provide disaster-recovery capability.

Figure 3. Sample Work Plan

8.15 Back Up and Disaster Plan
#15..
- 894 -
SOLICITATION # CH16012
156

8.15.1 Ability to apply legal retention periods and disposition by agency per purchasing entity policy and/or legal requirements.

CA APM Not currently available.

MAA Not Available.

CA Agile The customer is responsible for the lifecycle of their data. We will retain all data until it has been deleted from our systems. We can remove data from our systems upon written request from the customer at termination of the contract.

ASM Not available.

Salesforce Salesforce customers are responsible for complying with their company's data retention requirements in their use of the Salesforce Services. If a Salesforce customer must preserve data and the retention procedures above are insufficient, they may schedule a weekly export of data or copy to a sandbox account. Exports of Customer Data are available in comma separated value (.csv) format by request via Salesforce’s Customer Support department. In addition, many exports can be manually pulled by the designated org administrators.

ServiceNow Once notified ServiceNow has the ability to leverage our internal clone operation process. This provides a locked state copy of the instance, while the customer can continue working on the production instance. Any data additions, changes or deletes will be on the production but not on the frozen instance copy.

Additionally, ServiceNow can freeze backups to prevent them from being over written.

Docusign DocuSign customers determine their retention policies.

QTS Yes QTS will work with customers to meet legal requirements.

SAP Ariba For the network’s standard record retention is 180 days, but customers can request longer terms.

Fieldglass Fieldglass has many methods in which we can freeze customer data, today.

We have the capabilities to restore, utilizing a point in time recovery method to a secondary system for legal review effectively delivering a system that has been frozen in time for all clients.

Hanna HEC standard backup retention is 1 month for productive systems, and 14 days for non-productive systems. However retention period can be changed per request and may have cost impact.

SuccessFactors A customer can configure as many ""purge rules"" as required to comply with data privacy rules. Since these rules vary by country, the solution provides the flexibility to define different retention periods for groups defined by country, department, division, location, job code and hire date.

Customers can grant permissions to the appropriate users to create a request to purge data and to approve the request to purge data. These ""purge rules"" can be scheduled to reoccur. User records are kept in the system, for reporting purposes, but their data is anonymized. Our customers use the solution in compliance with local regulations today in almost 180 countries.

VMware VMware IaaS Services
Participating entities that leverage VMware Data Protection are able to specify the retention period of their backups to comply with policies and legal requirements. Data Protection leverages snapshots to take an image level backup of virtual machines. When a backup is initiated, a snapshot is taken of that virtual machine. Once the snapshot is taken it is then stored on the Data Protection storage for the duration of the retention period assigned in the policy. The policy column in the screenshot below illustrates the variety of data retention policies that have been set within a sample customer’s vCloud Air Data Protection instance. In this example retention policies range from 3 to 365 days.
#15..
- 895 -
SOLICITATION # CH16012
157
VMWare AirWatch
Customers can export and archive solution data using interactive dashboards (CSV), reports (PDF, XLS, and CSV), the AirWatch Hub (PDF), and event log (CSV). Customers in a Dedicated Cloud deployment can leverage a robust data mart.

FireEye Customers have access to data stored in the varying cloud solutions and are accessible at any time during the length of the subscription contract. Retention periods vary per service offered. See section 8.7.2 for further information.

VirtueStream Specific retention requirements can be accommodated on a custom basis. More detailed information would need to be provided in order to scope.

8.15.2 Describe any known inherent disaster recovery risks and provide potential mitigation strategies.

CA APM DR is provided only in a single geography at this point. Professional Services would need to be engaged to provide multi-geo DR

MAA N/A
CA Agile None - our DR plan is tested at a minimum semi-annually.

ASM None
Virtru Please see the Virtru Disaster Recovery Plan provided in the Supplemental Information section of this response.

Salesforce The Salesforce service performs near real-time replication at each data center and annual disaster recovery tests for the service verify the projected recovery times and data replication between the production data center and the disaster recovery center. The disaster recovery site is a 100% replica of the primary production site of capacity (host, network, storage, data). Data is transmitted between the primary and disaster recovery data centers across encrypted links. Additionally, backups of data are performed and data is retained on backups at the geographically separated disaster recovery data center location [CP-4, CP-6, CP-7, CP-9, MP-5].

ServiceNow ServiceNow is divided into two distinct environments for Business Continuity (BC) and Disaster Recovery (DR). ServiceNow’s corporate environment and its data center environments are physically and logically isolated from each other. A disaster in ServiceNow’s corporate environment could occur with little or no impact on the ability for the data centers within the private cloud to operate.

In both cases, the BC and the DR are supported by a series of SOPs that allow ServiceNow to quickly and effectively take action when it is required. #15..
- 896 -
SOLICITATION # CH16012
158
ServiceNow’s customer DR is documented in its Information System Contingency Plan (ISCP), which covers its data center environments, and includes all customer environments as well as the instances that ServiceNow uses to run its business.

ServiceNow formally tests this process on an annual basis and can produce compliance reports for customers requiring them. ServiceNow also uses the process of transferring instances to reduce the impact of maintenance on its service and will move instances out of one data center to the other on a daily basis. As a result, ServiceNow is very well practiced at the process of “failing over” customer instances.

ServiceNow’s BC process covers its functional offices and is a separate standard operating procedure from its customer environments. The BC Plan (BCP) has been developed in concert with the entire business including Business Impact Assessments (BIA) to understand the impact of the loss of any given systems or locations.

Docusign DocuSign is dedicated to providing the industry’s most secure eSignature solution through the DocuSign Security Assurance Program.

QTS Not having a current copy of your information.

Mitigation - QTS's DR High Availability Service provides continuous data replication of your physical and virtual server environments, ensuring you always have access to a current copy of your data, applications and operating systems.

SAP Ariba Ariba has a strong disaster recovery procedure, tested every 6 months though has never required to fail over to the backup data center. The RTO is 4 hours and RPO is 5 minutes. Please see the SaaS Technical Infrastructure Whitepaper for addiitonal details.

Fieldglass None.

Hanna SAP Hana Enterprise Cloud helps customers improve their Business Continuity requirements with High Availability (HA) , Disaster Recovery & Backup options.

High Availability (HA) - HA is achieved by failover to standby node Disaster Recovery (DR) - Systems are configured in the secondary datacenter.

DR is an optional service that can be included as per customer's requirements SuccessFactors We maintain an N+1 approach for all equipment in the hosted cloud environment, so that there is never a single point of failure. All customer database backups are encrypted and streamed in a secure manner from the customer’s “primary” data center to their “alternate” data center, allowing for a timely restoration of service in the event of disaster.

Based on the terms of the agreement, we will designate one of the data centers as the customer’s “primary” data center, with an additional data center designated as the “alternate” for data redundancy and disaster recovery purposes.

VMware VMware IaaS Services
The following table provides common inherent disaster recovery risks and potential mitigation strategies.

RISK Unmitigated
Probability
Mitigation Mitigated
Probability
Participating
;ｪWﾐI┞げゲﾉ;Iﾆﾗa ;
disaster recovery
plan leads to
High Risk Participating Entities are
encouraged to develop a
disaster recovery plan to
mitigate against service
No Risk
#15..
- 897 -
SOLICITATION # CH16012
159
service outage or
data loss in the
event of a
disaster.
outage or data loss. In
parallel, entities are
encouraged to leverage
vCloud Air data protection,
vCloud Air Disaster
Recovery, and other high
availability services while a
formal plan is written
Participating
entity chooses
inappropriate
data center
location leading to
service
degradation or
outage.

High Risk Participating entities are
encouraged to choose
datacenter(s) that is
geographically proximate to
their end users to maximize
performance. In addition,
location of disaster recovery
data center should be
geographically separated
from primary data center.

Low Risk
Participating
entity fails to
budget for and
acquire
appropriate DR
technology
leading to
inability to
recover from
disasters.

High Risk Participating entities are
encouraged to include DR
capabilities in their costs
estimates for cloud
deployments to ensure
budget is available to
acquire those services.

No Risk
Participating
entity fails to
ensure cloud
service SLAs (e.g.

RPO & RTO
objectives) meet
their mission
requirements
leading to
inadequate
service recovery
times and data
loss.

High Risk Participating entities are
encouraged to review SLAs
for each cloud service that
they are interested in
acquiring to ensure that
they meet mission
requirements. Care should
also be taken to ensure
service SLAs do not
significantly exceed mission
requirements to avoid
unnecessary cost.

No Risk
VMWare AirWatch
#15..
- 898 -
SOLICITATION # CH16012
160
Providing a secure and stable product that mitigates vulnerabilities, reduces risk and quickly responds to the evolving threat landscape is considered a top priority. Following the guidelines established in NIST SP 800-30, regular risk and vulnerability assessments are performed against the production environment to identify, assess and remediate emerging threats. When potential vulnerabilities are discovered, we follow a documented procedure to prioritize and deploy necessary patches.
•Vulnerabilities and threat-sources are identified by conducting interviews with various members of the Cloud Operations community and mapping data to the AirWatch business landscape. •We can provide additional information as conversations progress.

FireEye Varies by service and customer deployment. Will be defined during engagement setup.

VirtueStream Virtustream has built redundancy into all layers of the physical infrastructure in order to deliver SLAs up to 99.999% and mitigate any known risks. To further mitigate against the unknown, Core microVM’s are designed for mission-critical workloads and include reserved compute capacity in a secondary datacenter and automatic storage replication to facilitate the ability to restore operations in the event of an outage at the primary data center.

8.15.3 Describe the infrastructure that supports multiple data centers within the United States, each of which supports redundancy, failover capability, and the ability to run large scale applications independently in case one data center is lost.

CA APM AWS datacenters provide redundancy and failover within a single AWS EC2 zone (i.e., a single geographic location). Professional Services would need to be engaged to provide multi-geo DR
MAA MAA service is currently available from one data center only.

CA Agile We run a hot/warm data center configuration and have the ability to quickly failover if required. Our data centers are ~1300 miles apart to ensure redundancy. We perform full system planned switch over testing at a minimum semi-annually.

ASM ASM core services run in one primary data center, and can fail over to the DR site if there is a catastrophic failure.

AODocs As described elsewhere in this response, AODocs is built on Google Cloud Platform (AppEngine) and Altirnao does not manage this infrastructure.

Virtru We operate our infrastructure in multiple Availability Zones, and any given data center failure would not result in data loss.

Salesforce Customer Data for customers in Salesforce’s Government Cloud is stored in two of our U.S. data center locations. Our service is collocated in dedicated spaces at toptier data centers. Salesforce’s hardware is located inside of secure server rooms designated to Salesforce and separated by concrete walls from other data center tenants. Individual racks inside of the data center are secured with a lock. Specific racks are allocated for hardware supporting the Salesforce Government Cloud. Access to the racks supporting the Salesforce Government Cloud hardware is restricted to Qualified U.S. Citizens as described in the section above.

At the heart of this architecture, each customer application instance is supported by a multi-homed network configuration with multiple connections to the Internet. Production application servers are load balanced within each data center. Production database servers are replicated in near-real time to a peer data center within the same geographic region. #15..
- 899 -
SOLICITATION # CH16012
161
ServiceNow leverages this Advanced High Availability (AHA) architecture for customer production instances in several ways:
•In the event of the failure of one or more infrastructure components, service is restored by transferring the operation of customer instances associated with the failed components to the peer data center.
•Before executing required maintenance, ServiceNow can proactively transfer operation of customer instances impacted by the maintenance to the peer data center. The maintenance can then proceed without impacting service availability.

See the “ServiceNow Security, Operations, and Compliance White Paper” included with this response for more information.

Docusign Each component of our trusted platform - Hardware & Infrastructure, Systems & Operations, Applications & Access, and Transmission & Storage - undergoes tremendous security scrutiny.

HARDWARE & INFRASTRUCTURE
Three geo-dispersed, SSAE 16 audited datacenters
Near real-time secure data replication and encrypted archival 365x24x7 on-site security
Annual Business Continuity Planning (BCP) & Disaster Recovery (DR) testing Third-party penetration testing

SYSTEMS & OPERATIONS
Physically and logically separate networks
Two-factor, encrypted VPN access
Professional, commercial grade firewalls and border routers Distributed Denial of Service (DDoS) mitigation
Active monitoring and alerting

APPLICATIONS & ACCESS
Formal code reviews and vulnerability mitigation by third parties Application level Advanced Encryption Standard (AES) 256 bit encryption Key Management & Encryption Program
Enterprise-class malware protection
Digital audit trail
Multiple authentication mechanisms

TRANSMISSION & STORAGE
Secure, private SSL 256 bit viewing session
Anti-tampering controls
Signature verification of signing events
Unalterable, systematic capture of signing data
Digital certificate technology
Customer configurable data retention program
#15..
- 900 -
SOLICITATION # CH16012
162
QTS QTS cloud utilizes alternate telecommunications services from both TW Telecom and NTT America that reduces the likelihood of experiencing a single point of failure.

QTS's failover facilities in RIC1 and ATL1 are a carrier-class and carrier-neutral datacenter with multiple Internet Service Providers. QTS-FC can establish connectivity with another service provider with 24 hours in the event of an unrecoverable catastrophic connectivity failure of QTSFC’s primary/secondary provider.

Local Ariba staff maintains the hardware remotely.

Ariba maintains the application software.

Processes are in place to keep database and fileservers in sync between primary and backup data centers.

The failover process of all parts of the infrastructure is automated.

In the event of a catastrophe, Ariba will declare the primary data center “down” and locally the script will be run to switchover and start the applications at the remote data center.

Ariba tests power outage backup scenarios and the Disaster Recovery Plan on a periodic basis to ensure it is up-to-date, successful, and effective.

Fieldglass We host and manage the Fieldglass system within secure Internet data centers provided by CenturyLink in Elk Grove, IL, and our disaster recovery center provided by Equinix in San Jose, CA. Both are managed by Server Central which is a Tier 4 hosting provider and a Tier 1 data provider (see the note below).

All systems are ""mission-critical loads"" supported by IDC Redundant Power Management System (UPS, generators etc.).

Note: Data centers can be classified by tiers, with Tier 1 being the most basic and inexpensive, and Tier 4 being the most robust and costly. According to definitions from the Uptime Institute and the latest draft of TIA/EIA-942 (Telecommunications Infrastructure Standard for Data Centers), a Tier 1 data center is not required to have redundant power and cooling infrastructures. It needs only a lock for security and can tolerate up to 28.8 hours of downtime per year. In contrast, a Tier 4 data center must have redundant systems for power and cooling, with multiple distribution paths that are active and fault tolerant. Furthermore, access should be controlled with biometric readers and single-person entryways, gaseous fire suppression is required, the cabling infrastructure should have a redundant backbone, and the facility can permit no more than 0.4 hours of downtime per year.

Tier 1 or 2 is usually sufficient for enterprise data centers that primarily serve users within a corporation. Financial data centers are typically Tier 3 or 4 because they are critical to our economic stability and, therefore, must meet #15..
- 901 -
SOLICITATION # CH16012
163
higher standards set by our government. Public data centers that provide disaster recovery/backup services are also built to higher standards.

Hanna Infrastructure setup:
* Standby SAP HEC systems at DR failover site will be shutdown / not useable during regular operations.
* WAN connectivity & special requirements (e.g. acceleration services) fall into the responsibil-ity of the customer.
* Customer must have independent network connections to both DR sites. * All DR-relevant systems are going to be installed according to the Adaptive Computing (AC) principle (virtual hostnames + virtual IP addresses HSR will be used to replicate data from Primary to DR site for HANA solutions, similarly Sybase Replication Server (SRS) will be used to replicate data from Primary to DR site for ASE based solutions. For more information, please refer to attached Business Continuity document
SuccessFactors Our infrastructure architecture is designed with high availability in mind, and engineered for resiliency. All major components are redundant, including power, HVAC, fire suppression, and the physical components of our network.

Production data centers have strict access controls, and are continuously staffed and monitored to help prevent acts of sabotage or vandalism. All production data centers are ANSI/EIA/TIA-942Tier III/IV facilities, and are ISO 27001 certified.

Aspects of the plan are described in the SOC reports, Disaster Recovery Plan (“DRP”) and Business Continuity Plan (“BCP”) solutions are dependent on several factors. The customer may be responsible for parts of the recovery/continuity activities
VMware VMware IaaS Services
VMware offers several technologies to provide redundancy, failover capability and the ability to run large scale applications independently in case one data center is lost. vCloud Air is built upon infrastructure that is architected for High Availability leveraging proven vSphere High Availability (HA), vSphere vMotion, and VMware Distributed Resource Scheduler (DRS). These technologies allow participating entities to migrate live workloads and/or automatically restart VMs in the event of host maintenance or unexpected issues. These technologies also maintain a consistently high level of performance for all tenants.

VMWare vSphere vMotion enables migration of live VMs to transfer your workloads during maintenance, without downtime. Unlike other cloud providers, VMWare ensures that tenants are not impacted by standard maintenance. VMware performs vMotion migrations for maintenance behind the scenes for its tenants in vCloud Air during maintenance windows without disrupting tenant applications.
vMotion is also used to seamlessly migrate VMs from an on-premises environment to vCloud Air and back. Today, vCloud Air is one of only a few major cloud vendors that uses live migration to do maintenance. This capability will enable NASPO participating entities to perform live migrations to move thier applications to and from the cloud for maintenance and other purposes. This #15..
- 902 -
SOLICITATION # CH16012
164
technology inherently prevents the lock-in experienced when migrating to cloud services offered by others.

VMware Dynamic Resource Scheduler continuously monitors CPU and memory utilization across a cluster of vSphere hosts, allocating resources among VMs and rebalancing performance during high-volume peak times. This is performed 24x7 within the VMware vCloud Air environment.

VMware vSphere High Availability delivers the availability required by most applications running in VMs, independent of the operating system and application running in it. Whether participating entities leverage Microsoft Windows or Linux; off-the-shelf or custom developed application; single VMs or many, HA provides uniform, failover protection against hardware and operating system outages within vCloud Air.

HA can:
• Monitor VMware vSphere hosts and VMs to detect hardware and guest operating system failures.
• Restart VMs on other vSphere hosts in the cluster without manual intervention when a server outage is detected.
• Reduce application downtime by automatically restarting VMs upon detection of an operating system failure.

Other cloud providers require you to build availability in to your application. In the best case scenario, this is as simple as including a load balancer in front of multiple VMs in the same tier to achieve basic high availability.

However, often the technologies of other cloud service providers requires a complete redesign of the hosted application, and often network devices such as load balancers are offered at an additional cost, unlike vCloud Air.
vCloud Air was built to run hosted applications as-is, providing availability down to a single VM instance, and balancing resources across multiple tenant needs with DRS. Unlike other clouds, with vCloud Air and vCGS, participating entities are not required to rebuild thier application to get the flexibility and agility that cloud can provide. Of course, vCloud Air customers usually do not just stop at moving applications to vCloud Air. Once in the cloud they start to explore features like our Advanced Networking Services and using APIs, automation, and CI/CD tools to deliver services faster. And, where it makes sense, customers do look at application redesigns as well, but with vCloud Air it is not because they must redesign to ensure the application is always available.

VMWare AirWatch
AirWatch features active-passive configurations for high availability and redundancy with all components made to failover with minimal downtime. Load balancing capabilities are deployed across multiple data centers to ensure immediate server pick up, ensuring zero end user downtime. AirWatch also incorporates replication technology featuring SQL log shipping or network SAN byte replication to prevent data loss.

FireEye Varies by service and customer deployment. Will be defined during engagement setup.

VirtueStream Virtustream’s U.S. based data centers are located in Washington, DC, Las Vegas, and San Francisco. All data centers meet the highest industry standard in terms of operational protocol and have redundancy built into all layers of the physical infrastructure. Core microVM’s, which are used for mission-critical workloads, provide reserved compute capacity and automatic storage replication so that these systems can be failed over to a secondary data center and have access to the compute capacity required in the event of a catastrophic outage at the primary data center.

8.16 Solution Administration

8.16.1 Ability of the Purchasing Entity to fully manage identity and user accounts.

CA APM Customer has full control over all user accounts #15..
- 903 -
SOLICITATION # CH16012
165
MAA CA MAA provides user management self-service to customers to manage their own accounts and create accounts for other users.

CA Agile The customer is responsible for managing the lifecycle of all user accounts.

ASM The Purchasing Entity can manage their account and create sub-accounts within ASM.

Google Comply
AODocs Comply
Virtru Virtru does not have an identity system. We federate authentication to Oauth and Email loop verification.

Salesforce Identity Management
Logon is form-based. When users log into the Salesforce application, they submit a username and password, which are sent to Salesforce via an TLS-encrypted session. Security features are developed by Salesforce and built into the application. Third-party packages are not used for development or implementation of security internal to the application.

In addition, single sign-on and two-factor authentication may be used to authenticate users. Some organizations prefer to use an existing single sign-on capability to simplify and standardize their user authentication. You have two options to implement single sign-on—federated authentication using Security Assertion Markup Language (SAML) or delegated authentication.

Federated authentication using Security Assertion Markup Language (SAML) allows you to send authentication and authorization data between affiliated but unrelated Web services. This enables you to sign-on to Salesforce from a client application. Federated authentication using SAML is enabled by default for your organization.

Delegated authentication single sign-on enables you to integrate Salesforce with an authentication method that you choose. This enables you to integrate authentication with your LDAP (Lightweight Directory Access Protocol) server, or perform single sign-on by authenticating using a token instead of a password. You manage delegated authentication at the profile level, allowing some users to use delegated authentication, while other users continue to use their Salesforce-managed password. Delegated authentication is set by profile, not organization wide. You must request that this feature be enabled by Salesforce.

Salesforce can be configured to utilize Active Directory directly via Delegated Authentication, or indirectly via Federated Identity using either SAML 1.1, or SAML 2.0. Additionally your users can be loaded from information drawn from your Active Directory servers and modifications made in Active Directory can be propagated into Salesforce.

Customers can use their own SAML Identity Provider, or license one directly from Salesforce with our Identity Connect product.

User Accounts
All users and application-level security are defined and maintained by the organization administrator, and not by Salesforce. The organization administrator is appointed by the customer.

An organization's sharing model sets the default access that users have to each other's data.

There are four sharing models: Private, Public Read Only, Public Read/Write, and Public Read/Write/Transfer. There are also several sharing model elements: Profiles, Roles, Hierarchy, Record Types, Page Layouts, and Field Level security.

ServiceNow Customers manage their own instances, including assigning users, defining user roles and profiles, and administrating the system. ServiceNow does not have access to the Customer’s instances. The ServiceNow wiki has all ServiceNow documentation available. #15..
- 904 -
SOLICITATION # CH16012
166
Docusign Yes, DocuSign allows the purchasing entity to mange identities and user accounts.

DocuSign for Enterprise is customizable to meet the administrative needs of the largest organizations. Full administration rights give you total control over document custody and retention policies, how signers sign or adopt signatures, user authentication and more. The branding tools ensure recipients can easily identify documents from your organization.

QTS 24x7x365 Operations Staff - Let QTS provide the expertise and experienced staff to monitor and manage your data center facility. Never again worry about staff turnover or whether your technologists are current with data center best practices.

SAP Hanna Yes
VMware VMware IaaS Services
VMware management infrastructure requires 2-factor authentication and multiple layers of access control. vCloud Air customers have the ability to provide access to their designated users. User account management is maintained by the customer.
control. vCloud Air customers have the ability to provide access to their designated users. User account management is maintained by the customer.

VMWare AirWatch
The Purchasing entity has the ability to fully manage identity and user accounts for administrators and end users through the web console. Customers can optionally integrate with AD/LDAP to inherit your existing structure and policies.

FireEye User identity and account settings are controlled within the administration sections of the FireEye cloud solutions.

VirtueStream Virtustream can provide solution for managed solution for Active Directory or LDAP. As standard, we provide support for managing the VM itself and all OS support. However, our typical use case is, customer manages the users and related activities, as Virtustream might not be aware of internal HR elements of User Management.

8.16.2 Ability to provide anti-virus protection, for data stores.

CA APM AWS provides protection/virus scanning for data at rest MAA Logical security is provided by ClamAV Antivirus.

CA Agile We have ClamAV installed on all Linux hosts and TrendMicro for Windows hosts.

ASM Logical security is provided by ClamAV Antivirus.

Google Comply
AODocs N/A
Virtru AlienVault IDS is used to perform virus and vulnerability scans.

Salesforce Salesforce runs antivirus software on the production systems that store, transmit or process customer information. The Anti-virus scans host filesystems (not customer data). The antivirus software checks for virus definition updates daily. Other controls are also used to address malware such as hardening the Operating System of our servers, firewall configuration to ensure only required ports are open and all others denied, and use of intrusion detection systems. Access to these systems is restricted to authorized personnel and all these systems, as well as the host platforms, are monitored in real time through a security monitoring system.

The application only accepts http and https traffic, but Salesforce does not restrict the file types users can upload. Salesforce does not modify or clean any customer data; the system stores the information provided in an encoded format within the database. It is recommended that customers run updated antivirus and antimalware solutions to help mitigate these threats. The production system receives inbound mail as part of the workflow functionality, but this does not pose any threat to our network, application, or users. No code in the email can be executed or transferred, #15..
- 905 -
SOLICITATION # CH16012
167
eliminating the malicious software risk. Email sent from the Salesforce system is not currently scanned for viruses.

ServiceNow If Customers want to schedule an antivirus software scan on certain computers after a pre-defined number of days, Customer would define a model-based plan with a duration-based maintenance schedule. This would do the scan automatically to those models defined in the Planned Maintenance area.

Docusign Yes, DocuSign maintains enterprise-class malware protection.

QTS Yes - 24x7x365, dependable and detailed support
SAP Hanna Yes
VMware VMware IaaS Services
Anti-virus software is installed on all development, domain and production hardware devices owned and operated by VMware, which hold company data, passwords, or keys. These devices maintain the most recent version of the anti-virus software signature file.

VMWare AirWatch
AirWatch has installed a best-in-class, continuously-updated anti-virus suite on servers in the SaaS environment
FireEye FireEye employs a multi-layered, continuously monitored and engaged set of active defenses to provide anti-virus protection for customer data stores that both meet and exceed traditional approaches. FireEye, and our customers, not only are prepared to reactively withstand attacks from known virus vectors, but since FireEye and our MVX technology is literally, continuously, and securely detonating viruses and other advanced malware on the way to our customer’s data stores, FireEye proactively defends from zero day and other emergent attacks as they happen.

VirtueStream Virtustream offers Anti-Virus protection based on per virtual machine, per month charge.

8.16.3 Ability to migrate all Purchasing Entity data, metadata, and usage data to a successor Cloud Hosting solution provider.

Carahsoft will work with each Participating Entity to develop a cost effective plan to assist in the migration and implementation of their data into the new service that has been selected.

8.16.4 Ability to administer the solution in a distributed manner to different participating entities.

Carahsoft will view each Participating Entity as a separate customer and privileges can be assigned for control access even within the individual services.

8.16.5 Ability to apply a participating entity's defined administration polices in managing a solution.

Each Participating Entity will have the ability to work with Carahsoft and the Service Provider to identify defined administration policies to mangage the solution.

8.17 Hosting and Provisioning

8.17.1 Documented cloud hosting provisioning processes, and your defined/standard cloud provisioning stack.

CA APM CA SaaS Ops has a well defined Service Introduction and update process, including review boards, staged rollouts and followups to ensure service integrity. Our cloud provisioning stack is tied to AWS EC2 services, and includes S3 and RDS for storage; RHEL instances for the applications; route 53 DNS services; cloudfront for monitoring; and ELB for load balancing
#15..
- 906 -
SOLICITATION # CH16012
168
MAA Automation tools are used to install and configure MAA software on CA infrastructure.

SOP documents are used by CA internally.

CA Agile All new systems should be provisions according to our baseline standards and we use configuration management tools to ensure all systems are created with the same standards.

ASM Automation tools are used to install and configure ASM software on CA infrastructure.

SOP documents are used by CA internally.

Google Google has administrative document and Google Deployment Guides accessible on the web.

AODocs This is not applicable to Google.

Virtru Ubuntu AMIs provisioned using terraform and configured by Ansible Salesforce Cloud Hosting and Provisioning
Salesforce’s deployment model is a “public” cloud infrastructure, as defined by NIST 800-145. In the Salesforce Government Cloud, an agency dynamically provisions computing resources over the Internet on our multi-tenant infrastructure. This is a cost effective deployment model for agencies as it gives them the flexibility to procure only the computing resources they need and delivers all services with consistent availability, resiliency, security, and manageability.

Salesforce provides market leading PaaS and SaaS solutions and is a multitenant cloud-based subscription service. Multi-tenant cloud solutions provide a single, shared infrastructure, one code base, one platform that is all centrally managed, with platform-based API to support all integration traffic, and multiple release upgrades included as part of the subscription service. Multi-tenancy and the Cloud Computing model remove unneeded tasks from the process of delivering, managing, and integrating software. Salesforce customers will not need to maintain any hardware or software. Without multiple versions to support, integrations don’t break during updates; they are simply updated automatically. As a result, both the initial integration and its continued maintenance are simplified. More resources can be focused on creating a better product, with a faster cycle of innovation, instead of having to manage the complexity of many different versions to support a vast installed base.

Salesforce's position as an online service enables us to roll out all levels of improvement, from patch releases to major upgrades, that are largely transparent to the end users. When a bug is fixed and tested, it is rolled out to the application as part of regular maintenance; the nature of the service prevents special patches and code branches for individual customers, so all fixes can potentially benefit all customers.

The Salesforce Services is delivered using a world-class data center infrastructure. Each customer's org is hosted from a primary and secondary production data center, with near real-time replication occurring between the two sites.

Salesforce is a pure multi-tenant web application. No software or infrastructure is required by the customer other than a computer, browser and internet connection or a mobile device. User Administration and ProvisioningUser provisioning and management is performed by the customer through the Salesforce Administrative Setup environment. Users, their profiles, permissions and passwords may be managed, edited, activated and deactivated as needed by those with appropriate permissions. An administrator with appropriate privileges can manage session timeout, password policies, IP range login restrictions, delegated authentication/SSO, and requirements as part of this process. On first time login or password reset request, users are required to change their passwords to gain access.

ServiceNow ServiceNow includes all documentation on the wiki at wiki.servicenow.com.

Docusign DocuSign hosts customer accounts in its own data centers managed by DocuSign. Upon a new contract executed with a customer, DocuSign provisions a customer account based on the #15..
- 907 -
SOLICITATION # CH16012
169
contract terms. A designated point of contact becomes the default primary administrator who then adds additional users and other admin tasks. DocuSign's cloud stack is defined in architecture documents available under separate confidential disclosure.

QTS
SAP Ariba Our solutions are offered and delivered in a true subscription-based model and shared service (multi-tenant) offering. There is no software to install, no hardware to buy, no maintenance or support costs and no need to hire consultants or tech specialists to run the system. We deploy and manage the infrastructure. Customers only need a web browser for access. Subscriptions include system maintenance, automatic upgrades, enhancements and application of service packs, Level 1– 3 help desk support, professional services and best practices built directly into the application.

Fieldglass Fieldglass maintains its own platforms and follows a mature change control process for introducing new systems to the hosted environments.

Hanna The Managed Service includes the following:
• Infrastructure operations management: monitoring, patching, software updates & maintenance up to the OS Layer
• OS management: monitoring, patching, updates, and maintenance of the specific OS
• Network and system administration
• HANA database platform operations includes: space management revision management , security management , hardware configuration management, backup & recovery, change management coordination
• Health check services, proactive monitoring, capacity management • SAP Technical Application Basis Operations (incl. SAP Basis) • Monitoring
• Troubleshooting – Incident Management Level 2 and 3
• Patch Management
• Housekeeping
• Backup/Recovery
SuccessFactors We offer a cloud based solution. Our application is standardized on the J2EE technology stack with the majority of our software written in industry-standard software programming languages, such as Java.

We also make extensive use of Web 2.0 technologies, such as AJAX and Flash, for improved usability and performance and to deliver a rich and highly interactive experience. Our hardware consists primarily of industry standard web servers, application servers, database servers and storage and networking equipment.

Customers share the network security infrastructure, web servers, application servers and database instance, and each customer has their own set of database tables that are logically partitioned in the database containing its own unique database schema. Each company instance can be completely exported out of the database without affecting any other customer.

The application utilizes the JAVA and J2EE industry standards. Integration is supported through .CSV, XML and web services standards.

The application is portable to any platform and is deployed on Apache (Solaris) Web Servers, JBOSS Application Servers, and Veritas High Availability Cluster Servers, and utilizes Cisco Networking.

VMware VMware IaaS Services
An interface that allows users to select from a catalog of pre built OS images and allows them to provision on the fly.
#15..
- 908 -
SOLICITATION # CH16012
170
VMWare AirWatch
AirWatch uses industry recognized tools and follow best practices to deploy, configure, and maintain server images. We do not apply customer-supplied templates and do not share templates with customers.

Our Information Security team identifies applicable security standards and the AirWatch Server Team creates a secure baseline image for deployment throughout the SaaS environment.

Due to FOIA restrictions, additional information regarding capacity planning and image creation/safeguarding cannot be disclosed at this time.

FireEye This is not applicable as each of FireEye’s offerings are classified as SaaS.

8.17.2 Provide tool sets at minimum for:

1. Deploying new servers (determining configuration for both stand alone or part of an existing server farm, etc.)
CA APM New servers are predefined by the CA SaaS Ops toolchain, and automated using AWS tooling to be able to bring up/down servers on demand MAA Same as above.

CA Agile We utilize configuration management tools to ensure consistent configuration across our servers.

ASM Same as above.

Google 1. This is not applicable to Google. Capacity planning and server deployment is fully managed by Google.

2. This is not applicable. With the Google IaaS solutions all server images are provided by Google.

3. With respect to the SaaS offfering, Google Apps, if customers have not selected the Google Apps Unlimited option which offers unlimited storage, they have options to upgrade. With respect to the PaaS/IaaS offering, Google Cloud Platform, storage space is allocated dynamically based on actual usage demand.

4. SaaS, Google Apps includes numerous administrative tools to monitor end user and administrator activites."
AODocs This is not applicable.

Virtru AMIs provisioned using terraform and configured by Ansible to include all configuration parameters
Salesforce This is not applicable. These services while applicable to IaaS are not applicable to PaaS/SaaS where all of the infrastructure is managed by Salesforce as the Cloud Service Provider. These services are included and managed as part of the Salesforce subscription service and not directly exposed to the customer.

The multitenant architecture and secure logical controls address separation of customer data.

There are no dedicated servers used for individual customers. The Salesforce Services infrastructure is divided into a modular architecture based on “Instance”. Each Instance is capable of supporting several thousand customers in a secure and efficient manner. Services are grouped within each Instance; with app, search, and database elements contained. There are appropriate controls in place to ensure that any given customer’s org (application) is not compromised. The service has been designed to accomplish this and is robustly tested on an ongoing basis by both Salesforce and its customers.

An instance consists of a database instance run on clustered database servers and physical storage subsystems. Each server has been allocated a different volume group so it can fail over to its backup server within the instance independently of the others. As scalability needs dictate, #15..
- 909 -
SOLICITATION # CH16012
171
additional database instances can be brought online (the application already provides for this). For additional server capacity, one or more databases could be moved onto its own database server.

ServiceNow ServiceNow has read and understands this requirement. ServiceNow supports only a SaaS environment so there is no on-premise offering.

SAP Ariba Operating systems when installed through default means and with default settings are typically not secure and often include applications and services not required. Ariba has standardized operating systems' installations based on best practices. Ariba servers are configured with just the applications and services required to run the server as designed. Services not required are disabled and binary files not required for operation are removed, which reduces the total number of vulnerabilities a system may have.

Ariba standardized installation process also ensures that all servers of the same type are configured exactly the same, utilizing a process that initiates an upgrade to system binaries across all systems. Based on the system type, the configuration will remain the same. If one server type needs to receive a patch for a vulnerable binary file, that patch is pushed to all systems of that particular server type. If that binary exists on all systems, it is patched across all systems. Patches are then included in the build process as well so that future systems built are in compliance with security standards. Ariba uses this configuration process to 'harden' the operating system prior to implementing additional controls as detailed below.

Fieldglass Fieldglass uses group policy to ensure the configuration of all servers match our security requirements.

Hanna HEC is a private cloud offering where servers configurations are discussed and approved by customer prior to deployment.

SuccessFactors We apply hardening in line with ISO 27k standards and audit key areas in the SOC 2 Type 2 report. Audit reports are available to customers and prospects upon completion of an NDA.

VMware VMware IaaS Services
vCloud Hybrid Service provides an interface that allows users to select from a catalog of pre-built OS images and allows them to provision on the fly.
vRealize Automation is the next level for provisioning, delivering, and managing IT services on VMware vSphere infrastructure as well as VMware vCloud Air all with a unified management experience. It provides governance and blueprints to deliver services from on premise into vCloud Air.

Customers can integrate vCloud Air with vRealize Automation. In this setup, users access vCloud Air primarily through the vRealize Automation service catalog which handles authentication to the service itself. vRealize Automation supports
VMWare AirWatch
AirWatch uses industry recognized tools and follow best practices to deploy, configure, and maintain server images. We do not apply customer-supplied templates and do not share templates with customers.

Due to FOIA restrictions, additional information regarding capacity planning and image creation/safeguarding cannot be disclosed at this time.

FireEye This is not applicable as each of FireEye’s offerings are classified as SaaS.

VirtueStream Virtustream offers on-demand self-service portal where State of Utah can deploy new servers as stand-alone or part of a server farm.

2. Creating and storing server images for future multiple deployments CA APM Servers are brought up on demand
#15..
- 910 -
SOLICITATION # CH16012
172
MAA Same as above.

CA Agile We utilize configuration management tools to ensure consistent configuration across our servers.

ASM Same as above.

Virtru Provisioning system automatically persists all past AMIs for ability to roll-back Salesforce This is not applicable. These services while applicable to IaaS are not applicable to PaaS/SaaS where all of the infrastructure is managed by Salesforce as the Cloud Service Provider. These services are included and managed as part of the Salesforce subscription service and not directly exposed to the customer.

ServiceNow ServiceNow has read and understands this requirement. ServiceNow supports only a SaaS environment so there is no on-premise offering.

SAP Ariba Ariba has standard 'realms' which are default versions of the various configurations the Ariba on-demand products can be setup for each customer with one test and one production instance. Ariba maintains all hardware, software, realms and entitlements taking the burden off our customers and suppliers whilst providing a platform to configure to each company's needs and business processes
Fieldglass Fieldglass uses virtual servers within our hosted environments uses the hardening configurations to maintain the right level of security.

Hanna HEC is a private cloud offering where servers configurations are discussed and approved by customer prior to deployment.

VMware VMware IaaS Services
We also allow for template and gold master images to be stored on-premise and deployed to the cloud as well as storage of templates and master images in vCloud Director catalogs in in vCloud Air
VMWare AirWatch
Our Information Security team identifies applicable security standards and the AirWatch Server Team creates a secure baseline image for deployment throughout the SaaS environment.

Due to FOIA restrictions, additional information regarding capacity planning and image creation/safeguarding cannot be disclosed at this time.

FireEye This is not applicable as each of FireEye’s offerings are classified as SaaS.

VirtueStream Virtustream offers server images, templates, or other methods (OVF and blueprints) for multiple deployments as part of standard solution.

3. Securing additional storage space
CA APM AWS EC2 storage is elastic, and can be expanded on demand MAA Virtual storage can be added when needed.

CA Agile We have redundant systems for all critical infrastructure and have the ability to add additional storage if deemed necessary.

ASM Additional storage is added manually.

AODocs AODocs storage is based on Google Apps Drive storage and with respect to the SaaS offfering, Google Apps, if customers have not selected the Google Apps Unlimited option which offers unlimited storage, they have options to upgrade.

Virtru New hardware added at Cloudant
Salesforce Salesforce provides extensive storage capabilities. Storage is divided into two categories: file storage and data storage. File storage includes files in attachments, the Documents tab, the Files tab, the File field, Salesforce CRM Content, Chatter (including user photos), and Site.com assets.

Data storage includes the following entities/records stored within the Salesforce application:

Accounts, Article types, Article type translations, Campaigns, Campaign Members, Cases, Case #15..
- 911 -
SOLICITATION # CH16012
173
Teams, Contacts, Contracts, Custom objects, Email messages, Events, Forecast items, Google docs, Ideas, Leads, Notes, Opportunities, Opportunity Splits, Orders, Quotes, Quote Template Rich Text Data, Solutions, Tags: Tag applications, Tags: Unique tags, and Tasks.

For file storage, Unlimited Edition is allocated a per-user limit multiplied by the number of users in the organization plus an additional per-organization allocation. For example, an Unlimited Edition organization with 600 users receives 1,211 GB of file storage, or 2 GB per user multiplied by 600 users plus an additional 11 GB.

For data storage, Unlimited Edition is allocated either 1 GB or a per-user limit, whichever is greater. For example, an Unlimited Edition organization with 10 users receives 1 GB because 10 users multiplied by 20 MB per user is 200 MB, which is less than the 1 GB minimum.

Additional storage can be purchased, or files can be exported and archived outside of Salesforce, thus freeing up file storage space.

ServiceNow ServiceNow has read and understands this requirement. ServiceNow supports only a SaaS environment so there is no on-premise offering.

QTS Services that scale to meet your web and IT infrastructure needs - QTS offers space options from single cabinets to multi-rack cages to private suites. Coupled with configurable primary and redundant power options to run your infrastructure, QTS Colocation services can be scaled to support your near term growth requirements and future expansion.

SAP Ariba We carefully monitor all network interfaces. Internally we are well below capacity on our most congested interconnects. To the internet we use Border Gateway Protocol to peer with redundant Internet Service Provider’s and manage the bandwidth accordingly.

Fieldglass As a software-as-a-service (SaaS), Fieldglass automatically and transparently handles the scaling and redundancy for customers based on SLAs governing performance and availability. Fieldglass monitors application servers, database servers, bandwidth, and infrastructure for average and maximum utilization with Orion. Current utilization averages approximately 20 percent across all components with a goal of scaling the component vertically or horizontally (depending on the specific component) upon reaching a 60 percent threshold). All systems are scalable either horizontally or vertically to handle capacity requirements. This includes but is not limited to web servers, database servers, and storage.

Hanna HEC is a private cloud offering where servers configurations are discussed and approved by customer prior to deployment.

VMware VMware IaaS Services
Additional storage can be purchased and provisioned in 48 hours.

VMWare AirWatch
Our Information Security team identifies applicable security standards and the AirWatch Server Team creates a secure baseline image for deployment throughout the SaaS environment.

Due to FOIA restrictions, additional information regarding capacity planning and image creation/safeguarding cannot be disclosed at this time.

FireEye This is not applicable as each of FireEye’s offerings are classified as SaaS.

4. Monitoring tools for use by each jurisdiction’s authorized personnel – and this should ideally cover components of a public (respondent hosted) or hybrid cloud (including Participating entity resources).
#15..
- 912 -
SOLICITATION # CH16012
174
CA APM all SaaS based instances are monitored by CA SaaS Ops. With Hybrid deployments, customers can manage the on premise components of the deployment using their own tooling or CA products
MAA MAA is monitored internally as well as externally using multiple tools recognized within industry.

CA Agile Access to monitoring tools is provisions according to the principle of least priviledge and is only granted based upon business need.

ASM ASM uses Nimsoft Monitor, third-party open source monitoring tools, and ASM itself to monitor the health of ASM internal services and servers.

AODocs AODocs includes numerous administrative tools to monitor end user and administrator activites.

Virtru Dtrace, CloudWatch, CloudTrail, Logstache, ElasticSearch, Kibana, DataDog Salesforce Trust.salesforce.com is the Salesforce community’s home for real-time information on system performance and security. On this site you'll find:
ズ Live and historical data on system performance
ズ Up-to-the minute information on planned maintenance
ズ Phishing, malicious software, and social engineering threats ズ Best security practices for your organization
ズ Information on how we safeguard your data
System Overview
In addition to our Trust site (http://trust.salesforce.com/trust/status), you will also have access to a System Overview, which will help Salesforce customers monitor performance and usage of their own Salesforce org. This overview includes:

Schema - # and % of custom objects and data storage
Business Logic - # and % of Rules, Apex triggers and classes, as well as % of code used Licenses
API Usage - # and % of requests in the last 24 hours
User Interface - # and % of custom apps, sites, flows, custom tabs and pages Portal
The above list is of all the possible metrics that Salesforce customers may have in their system overview.

ServiceNow ServiceNow has read and understands this requirement. ServiceNow supports only a SaaS environment so there is no on-premise offering.

QTS Alert Logic’s Threat Manager Intrusion Detection System (IDS) and Vulnerability Scanning functionality are designed to monitor, detect, report and alert on adverse security issues on behalf of our customers. Within this process there are a number of both automated and manual processes which are supported by our SIEM software (‘The Expert System’) as well as our Network Security Analysts in our Security Operations Center (SOC). Threat Manager is the Alert Logic IDS and VA scan product. It includes the physical or virtual appliances and software tools used to analyze the customer’s infrastructure and monitor that infrastructure for attacks.

ActiveWatch is the managed aspect of Threat Manager, providing 24 x 7 monitoring and incident response operations from the Alert Logic SOC.

Threat Manager's Expert System identifies valid security events and suppresses false positives through a patented multifactor correlation process. Every security event monitored by its global network of threat sensors is analyzed in real time. When the Expert System determines that a set of events comprise a valid security threat, an incident is created and escalated according to severity via email or through an Alert Logic Network Security Analyst. This approach dramatically #15..
- 913 -
SOLICITATION # CH16012
175
reduces false positives and keeps analysts and customers focused on real, actionable incidents.

In addition, Alert Logic’s security research team continuously monitors threat data and tunes the Expert System to respond to the most current threats.

SAP Ariba Within our solution, specific logging takes place that is viewable by the customer administrator or their designee. Audit logs produced through use of the solution are considered customer data and are maintained within the customer's instance of the database. These logs are retained so long as the customer has an active contract with us.

As a user control consideration, customers are responsible for monitoring the proper entry of data to the solution and reviewing reports generated by the system.

Fieldglass Fieldglass has a comprehensive set of performance monitoring tools in place.

Some tools are built in the application and some are external.

The Fieldglass application captures every page hit and the time it takes to respond. It captures the server response time and the client rendering time separately. The average page times are available to be seen by privileged users through the application. The user can drill down to the hour or minute or a single request for analysis. The information can be displayed by time range for all users, by company, or even for an individual user.

Orion Network Monitoring watches more than 1,000 different points of interest on the production infrastructure. These monitors include, but are not limited to, the following: CPU, Memory, I/O, Capacity and Synthetic Transactions. These monitors along with our custom application performance monitors are used to produce and comply with customer SLA requirements.

Fieldglass also uses Splunk Enterprise Security for event/audit log collection and correlation. Events and logs are retained online for 90 days and offline for two years.

Alerts are generated and reviewed as they occur.

Hanna SAP HEC uses Solution Manager for monitoring the infrastructure end to end.

SuccessFactors The application logs the following for every transaction: Event/transaction Time, Transaction ID, Event/transaction Type, Event/transaction Status (Result of the event; if failure, includes reason), Object Attributes (Describes the object affected by the event), Originator User ID (ID of the user who initiated the event or action), Subject ID, Process User ID, Account Number, Transaction Specific Elements.

The application audits changes to the major components. Examples of this are goal auditing and document auditing. The audit trail includes who made the change, the date of the change and the ability to see the data as it existed at #15..
- 914 -
SOLICITATION # CH16012
176
that point in time. The data in the audit log can be viewed with appropriate permissions.

We also provide an optional Audit Framework for additional audit logging capabilities. All audit logs within the application are accessible only by customers. Therefore, the review of application audit logs as well as retention periods for the application audit logs are the customer’s responsibility, and can be determined as per requirements
VMware VMware IaaS Services
The vCloud Hybrid Service infrastructure, including the top layer management stack, the customer management stack and the computing/storage/network hardware are monitored for availability, capacity and performance. Customers are responsible for monitoring their own VMs (OS, apps).

Customers can use on premise monitoring tools to manage and monitor their applications and OSes over VPN and direct connect.

VMWare AirWatch
The AirWatch Cloud Operations team monitors the SaaS environment. AirWatch follows the Fault, Configuration, Accounting, Performance, and Security (FCAPS) model to monitor the SaaS environment.
•We have configured the system to notify support personnel of any issues with key performance items
•Because the solution is built on industry-standard ASP.NET architecture, it integrates into existing management and monitoring tools. Administrators can configure log data storage to destinations such as Windows Event Viewer, SNMP traps, syslog, SMTP email alerts, etc. The solution also integrates directly with Microsoft SCCM.

The communication layer includes a complete infrastructure for API integration to third parties as well as a rich set of existing APIs, web services, single sign on and authentication protocols.

Integrate with security information and event management (SIEM) solutions for enhanced logging of events occurring in the console. Administrators can view events, filter by event type, category and module, and export events. Event logging settings can be configured based on severity levels, with the ability to send specific levels to external system via syslog integration.

FireEye This is not applicable as each of FireEye’s offerings are classified as SaaS.

8.18 Trial and Testing Periods (Pre- and Post- Purchase)

8.18.1 Describe your testing and training periods that your offer for your service offerings.

Training and testing periods will vary based on the Service provider as well as the complexity of the request that is made by the Purchasing Entity. Some of the services can be functional within a few days and some will take a few months for full implementation.

8.18.2 Describe how you intend to provide a test and/or proof of concept environment for evaluation that verifies your ability to meet mandatory requirements.

Upon Request, Carahsoft will provide access to any of the Cloud Technologies proposed. The precise Access to the technology may be in a limited or “DEMO” fashion. Access will be to an extent that the State may complete any necessary tests or verifications that may be required. Instructions on precisely how each technology will be accessed will be provided upon request. All of the proposed technologies are currently live/ active and support many customers. A specific instance of the requested technology would be stood up within the cloud framework to support this request. As this will be occurring within a production, the creation of the instance is a quick and straight forward process not requiring any sort of extraordinary effort.
#15..
- 915 -
SOLICITATION # CH16012
177

8.18.3 Offeror must describe what training and support it provides at no additional cost.

Carahsoft provides the customer with free training for specific deals on a case by case basis. This can include free installation, training, and maintenance for high value customers who purchase Carahsoft’s solution through the NASPO contract vehicle. In addition, Carahsoft will provide webinars with basic training on a quarterly basis, made available to any Purchasing Entity who wishes to participate.

8.19 Integration and Customization

8.19.1 Describe how the Solutions you provide can be integrated to other complementary applications, and if you offer standard-based interface to enable additional integrations.

The integration into other applications will vary base on the Service Provider, however it is standard practice to allow for open AIs for integration development. For example Connecting Salesforce to an existing enterprise application is a common and frequently performed task. Integration options range from native Web Services support (APIs, outbound workflow, etc.) to import/export utilities to middleware integration via packaged connectors to toolkits for Java, .NET, and other open platforms. Our solution provides the ability to call out to virtually all common APIs, to enable synchronization, push / pull, and mash-ups with external apps/systems. Salesforce itself is based on web-service based APIs that in turn simplify access to Salesforce data from external systems. API-based integration is heavily leveraged by our customers.

8.19.2 Describe the ways to customize and personalize the Solutions you provide to meet the needs of specific Purchasing Entities.

The Service Providers customization options are very extensive and highly configurable. For example The Salesforce Platform offers a core set of technologies that not only power the Salesforce SaaS products, but also allows your Agency to build custom apps, connect data from any system, and manage it from anywhere. The Salesforce Platform allows customers to build apps fast with just a few clicks, designed for desktop and mobile devices, all from a single canvas. The Salesforce Platform has been given top ratings by Gartner, Forrester, & Info-Tech Research. To help IT deliver apps faster, the Salesforce Platform offers a simple yet powerful set of declarative, point-and-click tools that anyone can use to achieve business goals at lightning speed. Without writing code, developers and business users alike can quickly and easily create custom apps on the Salesforce Platform with complex business logic and beautiful user interfaces designed specific to every screen. Salesforce Lightning Builder tools allows your Agency to work in alignment with agile development methodologies as IT meet business demands faster. The Platform uses open APIs based on industry standards such as REST and SOAP to make it easy for your Agency to build apps that integrate with legacy systems. For more complex applications, developers can leverage the Apex programming language. Apex is an object-oriented, on-demand language. It is like Java, with similar syntax and notation, and is strongly-typed, compiled on demand, and fully integrated into the Platform. All of the application services come right out of the box, from a powerful workflow engine to API services, integration services, authentication, event log framework, analytics, and collaboration.

8.20 Marketing Plan
Describe your how you intend to market your Solutions to NASPO ValuePoint and Participating Entities.

Carahsoft offers deep experience in public sector marketing. Our dedicated team plans, promotes and executes more than 2,000 public-sector marketing campaigns and events each year, including contractspecific promotional activities. These include but are not limited to:
 News announcements
#15..
- 916 -
SOLICITATION # CH16012
178
 Social media promotion (Twitter, Linked In, Facebook, Carahsoft Community)  Website content/reciprocal links (Carahsoft website page; content for contract sponsor page)  Marketing materials (FAQs, contract overviews, solution spec sheets, powerpoint slides)  Training documents
 Co-branded tradeshow graphics, giveaways, display materials  Tradeshow participation (national, state and local government and education shows)  Digital and print ads
 Webinars
 Email campaigns
 Proactive marketing opportunity tar available through:
o National Coalition for Public Procurement (NCPP) – publicprocurementcoalition.org o Institute for Public Procurement (NIGP) – nigp.org
o National Association of Counties (NACo) – naco.org
o The United States Conference of Mayors – usmayors.org o National League of Cities – nlc.org
o National Governors Association – nga.org
o Relevant State Associations

8.21 Related Value-Added Services to Cloud Solutions
Describe the valued-added services that you can provide as part of an awarded contract, e.g. consulting services pre- and post- implementation. Offerors may detail professional services in the RFP limited to assisting offering activities with initial setup, training and access to the services.

In addition to Carahsoft's quoting and configuration expertise, we also have a wide range of Service Provider implementation partners that we leverage to work with Purchasing Entities in order to ensure successful adoption and deployment within a Purchasing Entity. We will help to identity needs for initial setup, training and access to the services so that it is all available at the best prices from one source to the Purchasing Entity.

8.22 Supporting Infrastructure

8.22.1 Describe what infrastructure is required by the Purchasing Entity to support your Solutions or deployment models.

The Service Providers that are being proposed operate on their own secure infrastructure. The Purchasing Entities will only need an internet connection to access the services.

8.22.2 If required, who will be responsible for installation of new infrastructure and who will incur those costs?
The Service Providers maintain all costs for updating and building new infrastructures.

8.23 Alignment of Cloud Computing Reference Architecture Clarify how your architecture compares to the NIST Cloud Computing Reference Architecture, in particular, to describe how they align with the three domains e.g. Infrastructure as a Service (IaaS), Software as a Service (SaaS), and Platform as a Service (PaaS).

CA APM The SaaS portal runs in Amazon Web Services.

MAA CA MAA is a SaaS service.

CA Agile Only SaaS is being offered via a public cloud
#15..
- 917 -
SOLICITATION # CH16012
179
ASM As a monitoring service, ASM is considered to be SaaS.

Virtru The capability provided to the consumer is to use the provider’s applications running on a cloud infrastructure2 . The applications are accessible from various client devices through either a thin client interface, such as a web browser (e.g., web-based email), or a program interface. The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited userspecific application configuration settings.

Salesforce Salesforce’s deployment model is a “public” cloud infrastructure, as defined by NIST 800-145. In the Salesforce Government Cloud, an agency dynamically provisions computing resources over the Internet on our multi-tenant infrastructure. This is a cost effective deployment model for agencies as it gives them the flexibility to procure only the computing resources they need and delivers all services with consistent availability, resiliency, security, and manageability.

Salesforce was the first Cloud Service Provider to attain FedRAMP Authority to Operate for both Software as a Service (SaaS) and Platform as a Service (PaaS), consistent with the FedRAMP moderate baseline controls. Salesforce does not provide IaaS as a direct service offering to our customers, it is an underlying part of our PaaS and SaaS offerings.

Salesforce Government Cloud
In May 23, 2014 Salesforce achieved a FedRAMP Agency Authority to Operate at the moderate impact level (as described in FIPS 199 and 200) issued by Health and Human Services (HHS) for the Salesforce Government Cloud. Additionally, on May 15, 2015, HHS, as the FedRAMP authorizing agency, approved the Salesforce Government Cloud authorization package that was updated based on annual attestation requirements and updates to the FedRAMP baseline which is FISMA compliant and based on the current release of NIST SP 800-53 Rev. 4.

Testing for the ATO was performed by a third party assessment organization (3PAO). The Salesforce Government Cloud information system and authorization boundary, is comprised of the Force.com Platform, Salesforce Services (Sales Cloud, Service Cloud, Chatter), and the backend infrastructure (servers, network devices, databases, storage arrays) that support the operations of these products, referred to as the General Support System (GSS).

To obtain compliance with FedRAMP, Salesforce conducted security assessment and authorization activities in accordance with FedRAMP guidance, NIST SP 800-37, and HHS guidance. As part of this process Salesforce documented a System Security Plan (SSP) for the Salesforce Government Cloud service offering. The SSP is developed in accordance with NIST SP 800-18, Guide for Developing Federal Information System Security Plans. The SSP identifies control implementations for the GSS and in-scope customer facing products (Force.com Platform, Salesforce Services) according to the FedRAMP moderate baseline and HHS security control parameters. A security assessment of the information system was conducted by a third party assessment organization (3PAO) in accordance with NIST 800-53A and FedRAMP requirements.

The security assessment testing determined the adequacy of the management, operational, and technical security controls used to protect the confidentiality, integrity, and availability of the Salesforce service and the customer data it stores, transmits and processes.

To maintain compliance with FedRAMP, Salesforce conducts continuous monitoring. Continuous monitoring includes ongoing technical vulnerability detection and remediation, remediation of open compliance related findings, and at least annual independent assessment of a selection of security controls by 3PAO. As part of our FedRAMP annual assessment, Salesforce is now aligned with NIST SP 800-53, Rev. 4 controls.

ServiceNow ServiceNow’s architecture aligns with Software as a Service (SaaS). #15..
- 918 -
SOLICITATION # CH16012
180
Docusign DocuSign’s DTM® solution is ISO 27001:2013 certified and many of the ISO 27001:2013 controls are mapped to the NIST 800-53 requirements; we can provide additional information upon request.

SAP Ariba Our solutions are offered and delivered in a true subscription-based model and shared service (multi-tenant) offering. There is no software to install, no hardware to buy, no maintenance or support costs and no need to hire consultants or tech specialists to run the system. We deploy and manage the infrastructure. Customers only need a web browser for access. Subscriptions include system maintenance, automatic upgrades, enhancements and application of service packs, Level 1– 3 help desk support, professional services and best practices built directly into the application.

Fieldglass Fieldglass is offered in a SaaS model. We follow the public cloud model identified by NIST. All customers access the same Fieldglass application version in a multi-tenancy database. The system is hosted in a secure hosting facility. Fieldglass manages everything from the cage in including all servers, devices, wiring, software, and configuration. Fieldglass uses the Cloud Security Alliance framework to measure itself against the recommended controls, in addition to ISO 27001 and SOC 2 controls in the Trust Services Principles of Security, Availability, Processing Integrity, Confidentiality, and Privacy.

Hanna HEC is a private cloud offering from SAP that provides fully 'managed' Infrastructure as a Service (IaaS) following NIST Cloud Computing Reference Architetcure.

Hybris The SAP Hybris Commerce, Cloud Edition is a Private Cloud Managed Service that offers a best practice multichannel commerce platform built on hybris accelerator technology, running in SAP Hybris own datacenters with a pay as you grow subscription model.

SuccessFactors Our IT architecture is aligned with ISO 27002. We are Safe Harbor certified and in alignment with BS10012 and ISO 20000 for Service Delivery. We demonstrate an on-going commitment to protecting the confidentiality, integrity and availability (“CIA”) of data from internal and external threats, making us a reliable and secure system provider.

Our secure multi-tenant Software as a Service (SaaS) platform is designed for availability, security, scalability, and performance. Industry best practices and standards are adopted and incorporated.

Our Network also complies with the Authority to Operate as a Moderate Risk Federal Information System by the Office of Personnel Management and Department of Homeland Security; NIST SP800-53 Security Controls; EU Privacy Directive 95/46/EC for EU and non-EU customer data; Payment Card Industry (PCI) Data Security Standard (DSS) v 2.0; Safe Harbor.

We provide privacy compliant data center facilities not only in the United States but also as a Member State of the European Union (EU) or a state of the European Economic Area (EEA). Currently, such data centers are certified for ISO27001, ISO9001 and PCI-DSS compliance.

Our security services provide complete and thorough monitoring of all traffic on the network on a 24x7x365 basis, and include security technology, alert services and incident management support. We are audited twice annually to SSAE16 (US) or ISAE 3402 (international) accounting standards. #15..
- 919 -
SOLICITATION # CH16012
181
VMware See section 8.1.1 for additional details.

Due to FOIA restrictions, we cannot share specific infrastructure details or architecture diagrams at this time. We can provide sample architecture diagrams under NDA to participating entities during task order negotiations as required.

VMWare AirWatch
AirWatch partners with co-located and cloud-hosted Tier III data centers to support of SaaS offering.

Due to FOIA restrictions, we cannot share specific infrastructure details or architecture diagrams at this time. We can provide sample deployments as under NDA to participating entities during task order negotiations as required.

FireEye Each of the FireEye cloud solutions are classified as Software as a Service (SaaS) and leverage a private cloud deployment model. With respect to the NIST Cloud Computing Reference Architecture, the following actors are integral to the overall design of the solutions:

Actor Definition
Cloud Consumer Customer organizations who have purchased FireEye cloud solutions
Cloud Provider FireEye
Cloud Carrier
Internet service providers and organizations
who have been contracted to host FireEye
cloud solutions.

A Cloud Broker is not leveraged in the FireEye cloud offerings. FireEye has contracted independent service auditors for evaluating controls relevant to security and confidentiality, however this occurs on an as-needed basis.

VirtueStream Virtustream solution is fully NIST compliant for Essential Characteristics, as the Infrastructure as a Service (IaaS) Service Model, with all deployment options – Private Cloud, Community Cloud, Public and Hybrid Cloud. To that extent:

1.Virtustream IaaS provides a self-service portal called xStream, where users can access, view, edit, provision, and modify compute, storage, network and application services based on granular Role Based Access Control, which can be integrated with Active Directory or LDAP.

2.Virtustream IaaS provides Broad Network Access, where it can provide landing zone for any private network (Point-to-Point, Virtual Private Label Switching, Multi-Protocol Label Switching, or Direct Connect), public network (Internet, Trusted Internet Connectivity as landing zone, IPSEC VPN and SSL VPN) and also extranet connectivity (shared network, i.e, Cloud Connect, Cloud Exchange, NetBond, etc.) which is growing rapidly as option for connecting cloud resources.

3.Key to Virtustream solution is its proprietary and patented solution of uVM technology, which is effectively a granular solution for resource pooling, providing application performance, pay for consumption only and segregate resources for security and compliance, but aggregate for cost efficiency. In addition, Virtustream is currently one of very few Cloud Service Provider with capabilities for Geo-Fencing and Geo-Tagging of the virtual machines to a specific data center, and getting down to cluster and host machine level.

4.Virtustream’s self-service portal, ticketing system and also the Technical Account Manager, who is the single point of contact for State of Utah allows for provisioning and deprovisioning of resources and services. In addition, Virtustream is uniquely positioned to provide application level provisioning as part of its standard automation and orchestration tool.

5.The fundamental of Virtustream solution is based on uVM Technology, which is very unique in terms of billing. Virtustream solution takes average consumption of compute resources, and State of Utah would only pay for actual resources based on a monthly average. This is in comparison #15..
- 920 -
SOLICITATION # CH16012
182
with traditional cloud solution, where the billing is based on allocation of resources in T-Shirt size (Micro, S, M, L, XL, XXL), and if the server is up, consumer of the cloud pays, but when it is down, they don’t. In case of the Virtustream solution, customer only pays based on the average of vCPU, RAM, IOPS and Network I/O; the key is average, not aggregate and based on consumption not allocation. In addition, all storage, security, application management services are offered as a monthly fee, based on the VMs; also, Virtustream can provide consultative and project support based on time and materials.

6.The deployment of Virtustream cloud can be on-premise or private based on purchase of its Cloud Management Platform, Community Cloud based on its IaaS, Public Cloud based on its IaaS and Hybrid cloud based on its ability to provide software and IaaS combined. #15..
- 921 -
SOLICITATION # CH16012
183
CONFIDENTIALが PROTECTEDが OR PROPRIETARY

INFORMATION
All confidential, protected or proprietary Information must be included in this section of proposal response.

Do not incorporate protected information throughout the Proposal. Rather, provide a reference in the proposal response directing Lead State to the specific area of this protected Information section.

If there is no protected information, write “None” in this section.

Failure to comply with this Section and Section 3.13 of the RFP releases the Lead State, NASPO ValuePoint, and Participating Entities from any obligation or liability arising from the inadvertent release of Offeror information.

Carahsoft Technology Corporation (“Carahsoft”) respectfully requests that section 6.3 Financials of Carahsoft’s response be treated as a trade secret under the Utah Uniform Trade Secrets Act, Utah Code ß 13-24-2 and submits this justification in accordance with Utah Code 63-G-2-305 demonstrating that its audited financials should be treated as a trade secret under Utah’s Uniform Trade Secrets Act. Additionally, Carahsoft is claiming the same confidentiality of the Consensus Assessments Initiative Questionnaire provided in relation to DocuSign and FireEye offerings, as well as the Cloud Controls Matrix for DocuSign.

Carahsoft has marked this attachment as “confidential” in accordance with the RFP instructions by including all confidential data into the Confidential, Protected, or Proprietary Information in our response.

Carahsoft is claiming confidentiality for these specific Consensus Assessments Initiative Questionnaires and Cloud Controls Matrix because releasing these documents would cause irreparable damage and would create the potential for cyber security breach. While portions of this information are in the public domain, the information contained within is not represented or assimilated this way in any other format.

Carahsoft’s Audited Financial Data satisfies the three-part test of the Utah Uniform Trade Secrets Act as set forth below and should therefore be treated as confidential and by the Utah Division of Purchasing and should not subject to public disclosure as provided in Utah Code ß 63G-2-305.

1. Carahsoft’s Audited Financial Data is a “compilation” within the meaning of Utah Code § 13-24-2(4).

Carahsoft’s Audited Financial Data is a compilation of all of the Company’s financial data and results for 2014.

While some aspects of this information may be available, the compilation of this information is unique and the result of sustained and distinct effort by the Company’s financial staff and its accountants.

2. As required by Utah Code ß 13-24-2(4)(a) Carahsoft derives independent economic value, actual or potential, from its Audited Financial Data not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use.

Carahsoft is a leading distributor of technology products and services to public sector customers such as the State of Utah, and throughout the United States and Canada. Sales in this market are intensely and deliberately competitive, with each sale subject to a structured bidding process in which low price is frequently the deciding factor on which contract award is based. Competition in this market is fierce, with competitors regularly competing with one another for federal, state, county and city business. #15..
- 922 -
SOLICITATION # CH16012
184
Carahsoft’s competitors closely monitor this market, and regularly attempt to gain financial and other information about Carahsoft and other competing vendors. Allowing Carahsoft’s competitors access to this compilation of financial data by releasing it in response to a public records request will damage Carahsoft’s competitive position because it will provide competitors insight into Carahsoft’s financial structure, including compensation, overhead costs, rental rates, profit margins and the like. Indeed, allowing insight into compensation may violate individual employees’ expectations of privacy. Once competitors have this access and insight, they can use it to undercut Carahsoft in the marketplace, damaging Carahsoft’s business with its customers and its suppliers, and inflicting economic hardship upon the Company.

For these reasons, Carahsoft does not itself disclose its Audited Financial Data except when required by law to do so. For example, Carahsoft discloses its financial information in filing its periodic tax returns—disclosure of which is prohibited by applicable state and federal law. In this regard, we note that Carahsoft is a privately owned company. Therefore, Carahsoft does not publicly report its financial results to the United States Securities and Exchange Commission.

3. Carahsoft uses reasonable efforts to maintain the secrecy of its Audited Financial Data.

The third part of the Utah Uniform Trade Secret Act’s test for trade secret treatment is whether the information “…is the subject of efforts that are reasonable under the circumstances to maintain its secrecy.” Utah Code § 13-24-2(4)(b).

Carahsoft satisfies this test because it takes reasonable steps to maintain the secrecy of its financial information.

As noted above, Carahsoft is a privately held company and therefore does not report its financial results publically to either its shareholders or the Securities and Exchange Commission. This is in contrast to public companies such as IBM or AT&T who report financial results quarterly as a matter of public record.

Even within Carahsoft’s business operations itself, Carahsoft’s management strictly limits disclosure of this information to a few carefully selected employees, and then only on a “need to know” basis in connection with their work assignments. Indeed, each Carahsoft employee signs a Confidentiality Agreement agreeing not to disclose this information.

Conclusion
Based on the forgoing, Carahsoft believes it has demonstrated that its Audited Financial Data, Consensus Assessments Initiative Questionnaires, and Cloud Controls Matrix are trade secret as defined in Utah Code ß 23- 24-2 . It is a compilation the content of which affords Carahsoft a competitive advantage. Carahsoft’s competition cannot readily discern it by proper means. Moreover, Carahsoft itself treats this information as a trade secret even within its own business operations.

Should there be a public records request for this information, Carahsoft respectfully requests that the State afford Carahsoft’s requested confidential information and documents trade secret protection in accordance with Utah law and deny the request. In the event that the State determines to release the requested confidential data, Carahsoft requests that you notify us promptly so that we may provide further justification for trade secret protection of this confidential data.
#15..
- 923 -

CLAIM OF BUSINESS CONFIDENTIALITY
Pursuant to Utah Code Annotated, Subsections 63G-2-305(1) and (2), and in accordance with Section 63G-2-309, Carahsoft Technology Corporation (company name) asserts a claim of business confidentiality to protect the following information submitted as part of this solicitation.

Pricing/Cost Proposals may not be classified as confidential or protected and will be considered public information. An entire proposal cannot be identified as “PROTECTED”, “CONFIDENTIAL” or “PROPRIETARY”.

Non-public financial statements
Specific employee name and contact information
Specific customer information, client lists, or subscription lists Other (specify): Consensus Assessments Initiative Questionnaires and Cloud Controls Matrix
This claim is asserted because this information requires protection as it includes:
trade secrets as defined in Utah Code Annotated Section 13-24-2 ("Trade secret" means information, including a formula, pattern, compilation, program, device, method, technique, or process, that: (a) derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by, other persons who can obtain economic value from its disclosure or use; and (b) is the subject of efforts that are reasonable under the circumstances to maintain its secrecy).
commercial information or non-individual financial information obtained from a person if: (a) disclosure of the information could reasonably be expected to result in unfair competitive injury to the person submitting the information or would impair the ability of the governmental entity to obtain necessary information in the future; [and] (b) the person submitting the information has a greater interest in prohibiting access than the public in obtaining access.

This statement of reasons supporting the claim of business confidentiality applies to the following information in this proposal:

Page Paragraph Reason
187-204,
Technical
Response
Template
All See Confidential, Trade Secret, & Proprietary Information Explanation

Page 176 ends

Modesto City CouncilJune 4, 2024

Modesto City Council
June 4, 2024