5/20/2025
tf
1 2
Introductions
Planning
Commission &
City Council
Meeting
Data Center Study Session
Item # 2, RTC 25-448
May 20, 2025
• Data Center Coalition: Dan Diorio, Senior Director, State Policy • City Managers Office: Jovan Grogan, Elizabeth Klotz
• City Attorney's Office: Glen Googins, Alexander Abbe
• Community Development: Afshan Hamid, Lesley Xavier, Steve Le, Meha Patel
• SVP: Nico Procos, Kathleen Hughes
• Economic Development: Reena Brilliot
• Water & Sewer: Gary Welling, Ahmed Aly
POST MEETING MATERIAL
Item# 2
1 5/20/2025
3 4
Purpose
• Inform community, Planning Commission and City Council about data centers
• Define a data center
• Provide background of data centers in Santa Clara
• Broader data center issues, covered by White Paper
• Opportunity for questions from Planning Commission, Council and public
Agenda
Background
What is a Data Center? Data Center Coalition
Data Center Land Use Regulations
SVP' s Power Supply and Power Planning for Data Centers Data Centers and Water
Economics of Data Centers
Questions
Item# 2
2 5/20/2025
5 6
Overview: What is a Data Center?
• Data centers are relied on as an essential infrastructure • Support the fast-paced growth of the high-tech industry and as necessary infrastructure to power cloud computing and now Artificial Intelligence (AI).
• Support significant growth in all AI in sectors: governments, businesses, medicine and households adopting digital transformation • Since pandemic, significant rise in digital transformation leading to increased demand for data centers, as AI requires substantial computational resources and storage.
• Average household has 21 connected devices
Dan Diorio, Senior Director, State Policy
Item# 2
3 5/20/2025
7 8
Data Centers:
Powering the Internet and Our Modern Economy
DATA~ CENTER
COALITION
What Do Data Center Providers Do?
Our members build, own, and
operate data centers
-----=--- -~--:-.~. :_ -= -..: ~~ ~ --
~---~ -• -"·
-- J._,-,/ 7 .
For their own operations,
one client,
or many clients
in a single building
Or for a single
company or client
on a campus
Item# 2
4 5/20/2025
Inside a Data Center
9 10
• Building Shell
• Interior Space
• Security
- Exterior
- Interior
- Cyber
• Servers
• Fiber/Networking Connectivity
• Reliable Power 24/7
- Grid & Backup Generation
• HVAC/Cooling
2 Main Types
of Data Centers
Self-Perform/ Enterprise
Business owns/controls servers and
peripherals, may own facility
Multitenant and Build to Suit
Facility owner leases to one or more
tenants
Item# 2
10 5
5/20/2025
11 12
Why Data
Centers?
Why Data
Centers?
• Significant driver of economy
• Enable digital infrastructure that supports our daily lives and modern economy
• Represent huge capital investments
• Generation of substantial tax revenue
• Build and support larger ecosystems of suppliers,
service providers, and other sectors of the economy
- Each direct job in the data center industry
supports more than six additional jobs
Item# 2
11 12
6 5/20/2025 Item# 2
U.S. Data Center Industry
Jobs
• 603,900 direct jobs in 2023-51 % increase from 2017
• 4.7 million in total employment in 2023-60% increase from 2017 • $404 billion in total labor income in 2023-93% increase from 2017 GDP Oo~ J'.
• $3.5 trillion in GDP impact between 2017-2023 t ij Ynl~ Taxes - Federal, State, and Local nn -nn
• $162.7 billion in total impact in 2023 - 146% increase from 2017 Soun:e: PH,,c, "Economic ContnbutJons or Data Centers m the United Stdtes, 2017-2023, "~ry 2025 13 California Data Center Industry
Jobs
• 2023 direct employment: 99,040
• 2023 total ( direct, indirect, and induced) employment: 560,450 Labor Income
• 2023 total ( direct, indirect, and induced) labor income: $65.8 billion GDP and Taxes Impact
• $122.9 billion (direct, indirect, and induced) to California GDP in 2023 • 24% increase since 2022
• $13.1 billion ( direct, indirect, and induced) in state and local tax revenues in 2023 Source: PwC, ''Economic Contributions of Data Cente,s in tlle United Sta/Es, 2017-2023, N February 2025 14 13
14 7
5/20/2025
15 ENERGY
Recalibrating global data center
energy-use estimates
Growth in energy use has slowed owing to efficiency gains that smart policies can help maintain in the near term
• In 2010, 79 percent of data center
computing was done in smaller traditional
computer centers, largely owned and run by
non-tech companies.
• By 2018, 89 percent of data center
computing took place in larger. utility-style
cloud data centers.
• While energy consumption by data centers
rose 6 percent from 2010 to 2018,
computing output jumped 550 percent.
Increasing Data Center Demand
• In the US market alone,
demand-measured by power
consumption to reflect the
number of servers a data
center can house-is expected
to reach 80 gigawatts (GW)
by 2030, up from 25 GW in
2024, according to McKinsey &
Company.
Global spending on the construction of data centers is forecast to reach $49 billion by 2030 .
Data center construction
&panding,1 $ billion
2022 2025 2030
Co-lc:c-o11io.,oa~
C\GR ~ 1'
• The United States accounts for
roughly 40 percent of the
global market.
,.,~ .,:,:•. L"'",I· ... :•= ""°'d'' :i ~-, c,, ~ .,~.,.,..L ~ ,, ,,)~~ ~~1 ...,r,• ~~ <l" • :., ar<l , , 1~-,, : ,~,, a ~•:.'<'' ''3 ~ ,li,.:l<',;; :;r,;1n,,,.,_,,, •~-Ch ..l· (";j1; r •.•:•' -., c..: ~v.,r1'l) F,.r•l·.'l':n; ~ ~
Md<lnuy4~
SaVr(e: McKmsey & Company, 'How data centers and the energy sector can sate Al's hunger f'or power", September 17, 20.24, /!l~~™'-' - -~~~~-~~~ 16
16 Item# 2
8 5/20/2025
What Drives Data Center Demand?
Global data created annually in zettabytes
23%CAGR
2024--2030
500
375
250
I I
125
11iJ 0
-
~
~
~ ~ :!; ~ !cl ~ ~ ~ 0 ~
f::j ::::i ~ ~ "' .... <X> m g ~ N N N N N N
0 0 0
~
0 0 ~ 0 0 0 0
~
0 0 0 0
~
0 0 0
~ N N N N N N N N N N N N N N N N
S,ourcr:J LlR-..-irch IOC
Source: Urban l and, Nuclear Po1Yer /ilakes a Comeb.::Jck as Data Ce11ters Adapt to Rising Power Demands; October 1 !, 202 ./, Qt!ps;M,. ntlntf,tJII.OIO/§YWCC"flnd•OJrif,P.tfLIP&',{11K/@t P2H:S::,. ~Nt-d.fU::f?J!!'M:1'14R( {O::ftitp •@ 1't!r-dclnlJl'lfll 17 Number of People/Devices Drives Data Center Demand
r ~
"The data center industry has experienced explosive growth over the past decade, driven by ever- increasing demand for cloud services and the expanding use of web-enabled devices globally. [ ... ] In the next five years, consumers and businesses will generate twice as much data as all the data created over the past 10 years."
'" -JLL, Data Centers 2024 Global Outlook ~
More People Are Getting Online
• Approximately 5.4 billion people - or 67% of the global population -are online today. This represents an increase of 45% since 2018. 2.6 billion people are not yet connected to the internet. • On average, U.S. households have a total of 21 connected devices.
Sources: JLL, Data Centers 2024 Global Outlook. b1t0$,·11:m;tr,u&iW.f911h.mrt:n@m!J(l-fll'lYillOrmc!rlt,'Pt(h;§Oyt'fMt'\?h.t{M:4-tt,t"frGl'te9utJ(J(lt•gbD,Jl•){l)-l,ptl lntemational Telecommu,1lcatio11 un,on, 11r,pr.-/Jwww.J1u.k1(harrtu ,OIStJtrUi'Jl&nrs,(gpf/df:l,1lJftr4w, Item# 2
17 Deloitte "Shmy new devices may oe brmqmg j oy, but wlto 's protectmg consumer data?", January 1), 102) h(lf-"F!/Ah1W}.dt;.bi©C?nOtSC-rot'tAAslDal~r'~•9'.JIJ ·P'l'VlQ' html 18 18 9
5/20/2025
a,
C,
Home Internet Use Drives Data Center Demand
700
600
500
400
300
200
100
0 2010
Average Monthly Household Broadband
Consumption
2012 2014 2016 2018 2020 2022 2023
Sour(~ 01)<:nVauft arc;d~nct Jns,gnrs R~rt, DVB! Q~ 203_~ R~rr, nmx:14»:rnM~AI 19 '
New Products/ Experiences/ Applications Drive Demand
• Cloud • Virtual/ Augmented Reality
• Generative AI • eCommerce
• Business Apps
• Healthcare • Machine Learning
• Internet of Things/Connected Devices • Payment Processing • Streaming Video • Online Learning
• Autonomous Vehicles
• Innovation!
20 Item# 2
19 20
10 5/20/2025
Growth in Data Center Construction Spending
$311.000
PS,000
~ l t,,000
$10,000
$:J,000
so
Value of Data Center PiP Construction
Spending
(Seasonally Adjusted Annual Rate, in $Millions)
.... V '<J' LO ,o '° •.o 'D :.c;. ,...... ..._ ,..._ :>.') ::r, ro ·:n a; (ft ;::-, C".) ,: ; ' ,- ,- ·'\j :"'sl 0,1 _..., ,:,') ._...., ... ·-t - ---- - ------------ N ~NNNN N NNNNNNN -::: > ~C >-6.::: ~a.::: :.'....~C >,,;_c: .;..QC>-~::::: :t..6...:= >-::l.~ >a.-:::>-;um~ m ~ ~ ~ m ~ ~ ~ m ~ ~ ~ ~ rn m w ro ~ ~ ~ ~ ~ ~ m ~ ro ~ ro ro ~ 2~~~m~2~~~~~~~~ ~ w~~m~~~~~w~~w~~
- Data center conslruct,on spe: nd1nq
Soutce: U.S. Census Bu/mu COIIStrodion Spendin!J DatiJ: HiStrJrical Value of Prwate ConstnxtiOn Put ,n Place (PiP}, May 2024 21 Data Center Trends
CHANGE IN U.S. CONSTRUCllON SPENDING, MAY 2023-MA Y 2024 The year•over-year percentage change in US. const rucllon spending in current dollars. seasonally adjusted .
To1llleons lluct;0n
..,.
Pl'N.1terevden1.111I ....
Nonr.-..denOI I ...
01t.ac enl8"1
~lactum,g
""'
~ (oncludi ng OIi and 911)
-""
Highway arad lttttl ...
"'~- .,,.
HHllh~ re
I "
r,..,u,o,1auon la
Ccw,unl'foal (w1mrhouw , retail, l ■ rm)
-
Olfooe(• .. clud, ngd.11acen1 e1,) ., ... ~
Source As.socia1KI ~ Cont~IOfS ol Amei-icai July ?024 rl'p()r1 Source Atlanta Susmess ChrOl7lCte, 518 dara centers adds to growm9 l1st of masswe pro1ects; offsets canstruc:,on s101wio1'111, Oct. -I, 102.J, C.~~{l'mt,J'.,.,~~~ ror-cower,,-cpunry.htm/
22 Item# 2
21 22
11 5/20/2025
23 Key Siting
Considerations
Include
• Access to Fiber/Interconnection
• Access to Water for Industrial
Purposes
• Access to Clean, Reliable, Affordable
Energy
• Climate and Risk of Natural Disaster
• Land Availability and Cost
• Tax and Regulatory Climate
• Ownership/Occupancy Costs
• Time to Market
• Access to Skilled Construction and
Technology Workforce
Global Data Center Markets (By Power Capacity)
24 Top Markets by Operational IT Load
5,000
4,500
4,000
3,500
3,000
f 2.500
2,000
1,500
1,000
500
11111111111111 11 1 1 ■
■ Americas ■ APAC EMEA
■
Item# 2
23 24
12 5/20/2025 Item# 2
Data Center Trends
User Demand by Industry Varies Across Markets
25 Northen California Pacific Northwest Phoenix
Cloud
• Technology
Telecom
•
Healthcare
Banking& Financial Services
• Retail & E-commerce
Entertainment & Media
• Energy
Other
Data Center Trends
Figure 6: Total Inventory vs. Under Construction by Primary Market, H2 2024 Megawatts (MW)
3500
e H2. 2024 Total Inventory e H2 2024 Undef Com:truc:tion Source: CBRE R8588rch, CBRE Data Center Solutions, H2 2024.
Source • ..,8RC. ·/orlll Amer .:J ;)d/iJ ·'.:!mer l"rtY1:ds 112 zo;: I, Fab.ruary 26, 202S, /U!Jlttltn1•1r.cbm.02tnl,ru;"7bl}ifo:emt@rflHtmg('d::t;UlJ «arcc·!t&!111·/tk?QJ.I 26 25
26 13
5/20/2025
Data Center Trends
Figure 3: X of Total Primary Market Inventory
Sourm:: CBRE Rmaard\ CBRE Data Center Solutions. H2 2024. • Northern Virginia
• Dallas-Fl. Worth
• Silicon Valley
Chicago
• Phoenix
e Atlanta
• HIiisboro
• New York Tri
State
'iolirCC =sRE, PfrWIII)(, NorJfl Amcoca Data C~l1ter Trends 112 202-.,, February 26, 2025, /ulM.'UNo»r-cfNttpm/ThCJa(Ut/lqg/•tCfOlWWtMttt>:Qmc,rtt1{i1':f'PIU~l ·rrettifi·hl·J<1cl awffl ·qn,[fftW(1 27 Data Center Trends
Figure 7: Total Inventory vs. Under Construction by Secondary Market, H2 2024 Megawatts (MW)
500
129
SOu1htrn
C,hfott1la
8.5 Se:Htlt
e H2 2024 Tott I Inventory e H2 2024 Under Construction Source: CBRE Research. CBRE Date Center Solutions,. H2 2024.
JIJu«e. :8RC, North Amc•.e.,ca Data t:.C1Ucr Trc11ds .'fl !02.J. F~bruary l6. 202!-. hrtrt1:•(l;,mr (1/(q cpmlfr#l(rfShtlKMS.IPPttfNWCt&HHYt«Of!V':frcnd'f•t,z,ztJl:J 28 Item# 2
27 28
14 5/20/2025
29 1800
1600
1400
1200
1000
800
600
400
200
000
Data Center Trends
Figure 3: % of Total Secondary Market Inventory
• Central Washington
• Austin/ San
Antonio
• Southern California
Seattle
8 Houston
• Denver
• Minneapolis
• Charlotte/ Raleigh
Source CBRE Re'iearch, CBRE Data Ctmt~ So/ut1oris. H2 2024, . :;, f' rrr,;Jl.t. V 1~ an:«rn:'<?i:9"0ic"k<tK:010:Yl!:e·M#'l:«nmritM◄~ Data Center Trends
Absorption Outside Top 10 Markets NA (MW)
- - - ■ 2018 2019 2020 2021 2022 2023 2024
Source: datace11terHawk, OltM•/lqi)bl:Ct:rm:'tM'i'ik com/ 30 Item# 2
29 15
5/20/2025
31 32
Afshan Hamid, Community Development Director
Lesley Xavier, Planning Manager
Steve Le, Senior Planner
Meha Patel, Associate Planner
Background
• City of Santa Clara home to high-tech companies
• Power-house for high-tech companies
• Since 1996, data centers supported fast-paced growth ofhigh tech, cloud computing and AI
Item# 2
32 16
5/20/2025
Background - Why Santa Clara?
• Access to Fiber /Interconnection
• Proximity to Clients and High-tech Work Force
• Access to Silicon Valley Power
• Certainty of review, development friendly environment for Entitlements
• Access to Recycled Water for Industrial Purposes
• Availability of industrial sites
33 Background
Santa Clara is home to:
• 57 active or under
construction stand
alone data centers,
• 1 entitled (not constructed)
data center
.....
I
·---
\
-
33 • 1 active planning
application
Approved I Built / Under Construction
Approved I Unbuilt
Pending Approval
34 34
Item# 2
17 5/20/2025
35 36
Nico Procos, Interim SVP Director
Kathleen Hughes, Assistant Director
Santa Clara's Electric Department
1967
1896
Cl.-a f1«1ric
[1985 "
I. ::::!=: by ,.
Altamont Pass, the
1998
$ill!,()n
VaJieY __ 1'17-Net 2013
I. SVP
pnnnde. «~
fr<•
: cityu,;J,
2021
: SVPcelebrat es ic.,
2018 125th y,aro/
S VP
1. seniice/orSanla
delivers ~ Clara residents.
dean '. 111<"0~
CU~o/Santo 1 ·
t>t,K,rt,n:'flr Cl
(OU'1~1l!hc n ll
dio:itl pe;k'l'DW,.
.Uarf.itpi,U;-trl,tg •
H1Jdro power is
aJJcx:aledfrom the
Fttlero/ C,,,trn/
Valley Pro;«t ,
beginningt.lte
di.,,..,;lcwi<m •f
: Cily'!first :
; cvrl,c,11-/n.y-rnau,a- !
: asitk/rom '.
TheCity'sE1ectn 'e
Departmetit
b«omt"S Silicon
Valley Pow,r (SVP)
tore/feet ils
growingro~.
ovldoor
Wi-Filll"Ct'SS
100,. ~ ~
fflffVIICOresidenlS. ~ V Ai if
~y"PcMK.
.J:i1rrn:tliJ,liu. : r"eSOIIJ"CeS.
+ 1903 _,_
~ c m ~
Santa Clara lllr,ests
$5,000 in
altemaririg cummt
and begins providing
electricity to
mstonwrs within
Santa Clara.
hvdroekctric poWl!T.
1980
Cogen I is laurrhed.
t.hit. Cifll'sfirst
in-cir,, power plant,
o natural
r,a,-powered
facilitt,.
198 ! 1' 2005 _ ,_.
Gn>:hmnol pott\·r ! The Donald Von
/ 10111t/rrN-artlr8u11 · Roe$/eldPower
An:a •oddnlro,Jw Plant , lo:atedin
Cat51·•1 """"'ti-Ir• SantaClaracomes
Ottkr toll'tlJl\a\"' on lineJroi,·iding
rri i.abllir11c11 d ,taturalgasand 147
rompr t h~·,- megau'attsof
priet""g. t?leetn"citytoSVP
cu&tornn-s.
2017 2020
i
SVPditJCSts its
i
SVPlaun:MS
ownership in th,e Electric Vehicle
San JuanCoa.l publicandjlttt
Pltmt . infrastnictMre
investment .
Item# 2
36 18
5/20/2025 Item# 2
37 Why Santa Clara? A History
• In 2000 SVP had requests
to double load ~ 450 MW
to900MW
• SVP added DVR power
plant & 230 kV
transmission line
• Downturns of the 2000s
• Economic Development -
Promote what makes Santa
Clara better
• City strike team that lead
economic development
NOT YOUR TYPICAL BAY AREA CITY
LOWER OPERATING COSl5
• l..ou'l'St combined urili[)' rJICS
of :iny &)' Acea city
• Snings of JQ.-«)% on d«Triciry
from 5.:lnu Clara's Silicon Valley l'owcr
• Nr:> utility raxcs
• RJtcd one of rnun1ry·s mO)t
;1fforihblc dri~ for business
{1004 Kosmom-Rosc Sun·cyl
FINANCIAL STABILITV
• Cit)"s annual operat ing l>udgcc
conrinu.tlly displ:1~ fi1c.1I
respomib1lirr wh1k mainraimn g
the highcu lc,;cls of scrv,cc
• Low business taxes, solid re.JI csrntc
YJluc anJ low nimc raCC'S ..:oncnburC'
to the Cicy's economic cfficicnq
.1nd prcsr,cricy
DEPENDABI£ AND AVAILABLE
INFRASTRUCTURE
• El«Trk rtli.abiliry conslSC'nuly r-.1n.1':s
in rhc top quanilc of lhc nation
• f:lmric, Water anJ Sewer capac iry
in place anJ available
• City's Fiber Opric Network conna:ts
ro c:Yl"f}' nujor Point of[>~ tPOPJ
• Nitrogen pipeline offtn un tquc
:tcces~ in S:1nr.i Cl.ar:a
PUBLIC SAFETY AND SUPPORT
• OursranJing Public: S:afety record
\\ith .1 Cl.1ss 1 Fire Rcspon5ie r .1ri'lk;
:tlso r,1nL:cd one of tht safC'Sf U.S.
1..JtttS wit h ;a 75.000+ popul.u ion
• Affordable housing :1nd ncittiJborhood
impro,·cmcm pmg.r::ims support our
highly eJuc::itcd, rr.1.incd :rnd diYcrsc
rmploymcnr rool
From Economic Development Brochure - Marketing Material 2003 SPEED TO MARKET ATTITlJDE
• Streamlined Busincu Pcrmir Program
offcn O,·cr-r~Counccr ~views • .1
Single Poim Con t:aa Sysrcm :ind fast
Tr.id:. Pcrmining dtsi~ncd to quickly
put your busineu on the map
• One-5top pa-mining cc:nrct hanJlc,;
most business services under ont
roof 1dw:rc 90o/ .. of rhc pcnnirs 3rt'
issued ovcr-lhc-<ountcr
• F..conom-= dcvdopmc nr rcpr~mJt i\'ts
3nJ procedures min imiu 3J)PrD\'.Jls
,1nd.dcl:ays
General Fund Contributions
37 • In the City of Santa Clara, data centers contribute 5% of SVP utility fees paid toward the City's General Fund.
• Each Megawatt (MW) of growth of usage from SVP adds approximately $52,000 a year to the General Fund. As such a 50 MW data center could contribute up to $2.6 million and a 99 MW could contribute over $5.0 millio annually to the City's General Fund.
• Significant future General Fund contributions based on forecasted growth:
Actual Budget Forecast Forecast Forecast Forecast
Description FY 2024 FY 2025 FY 2026 FY 2026 FY 2027 FY 2028 GF 5% Contribution $ 33.1 M $ 38.9M $ 41 .4M $ 44.3M $ 48.1M $ 51.7M 38 38
19 5/20/2025
39 40
Afshan Hamid , Community Development Director
Lesley Xavier, Planning Manager
Steve Le, Senior Planner
Meha Patel, Associate Planner
Land Use Regulations
• Data centers are allowed in:
- HJ - Heavy Industrial
- HO-RD - High Intensity Office
D U - Light Indu strial
=:J LO-RD - Low Intensity Office
• Data centers are not allowed in:
:----7 Rl
:=1R2
R3
R4
- RS - TN CC - Community Commen:lal
=1 CN - Commerdal Neighborhood
- C-R • Community Regional
~
• 1'kJ·NC - Mixed lJ9e Neighborhood Commemal
- MU·RC · MiJEd lJ9e Regional Commen:ial
- MU·VHD - Mixed lJ9e Very High Denoily Res.
OS - Open Space
- PQP - Public/ Quasi Public
UC-Urban c.e,,12,
-=:J IN· Urban Vdlage
VR - Village Residential
- Downtown Fonn Based
~ UC-ED
40 Item# 2
20 5/20/2025
41 43
Land Use Regulations
Land use districts ~~ ~- - ~ f
where data centers
are allowed near
those where data
centers are not
allowed.
HI - Heavy Industr ial
HO-RD - High Intensity Office
["7 U - Light Indu!.1:rial
D LO·RD - Low Intensity Office
Railroad Line
• Data Center Locations 41
Development Standards
Updated Zoning Code
(2024)
The Zoning Code provides for
specific development
standards for data centers in
addition to the standards such
as building setbacks and
heights.
Developmenta l Standards. The following standards shall apply to all Data Centers :
1. Primary Structure Facades. Primary structure facades shall include all structure facades that face adjacen t public roa<!s Primary structu res may house a primary use. and secondary (fr.,.,stand/ng) structures may hou,;o the data centor . or the data center may bo included In the prima ry structure. Pnma, y structure facades assoela!Cd with new eonstructlon shall meet all of the following standards:
._ Prlma,y stnJCturc rnl"'d• design shall nvoid Iha u~ of undff!e<enllated surfoces by Including al leaot two of the following design elements; change n strueruro height. struchJre step-backs or recesses, fenestration, change in structure material, pattern, texture , coklr , or use of accent materia ls.
b. When a primary structure has more than one facade, the facades shall be consistent in tenns of design. materials, details, and treatment.
2. Exte rior Lig hting. All e~eijo r lighting shall bo desig ned one! constructed wilh culoff ond M ly shielded fixtures that diree1 lighl downwa!d and into the nlerlor of lhe pl()perty and away from adjacent roads and properties in compliance with Section 18.40.080 (Ouldoor Lighting) .
3. Structural Noise Reduction Meaures . All data cenlers are required to provide structural noise reduction measures for any exterior cooling fans or equipment, such as baffles or acoustic louvers. to the satisfaction of the Director of Community Development. 43 Item# 2
21 5/20/2025
44 45
Development Standards
Updated Zoning Code (2024)
Off Scott Blvd Off Scott Blvd
Development Sta
Updated Zoning Code (2024)
Major Tech Company on Mission College Blvd Mission College Blvd Item# 2
44 45
22 5/20/2025 Item# 2
Land Use Images
Other Industrial Uses vs Data Center
--New Office
46 47
Land Use Images
Other Industrial Uses vs Data Center
Data Center
-
. .,. ' ' ~ . It . • •
I~. ·, --~~:._ --~- ..,...- •. ---~~-.
~ __ :,., ~ . ----- ~- - - --.. - . ..,.~ ~. ·•·:-··- -::._:_··-:·- New office with manufacturing building
New Office at Lawrence Station
,,_.,,,,,--~ • -~ --=--
Approved Industrial Building
46 47
23 5/20/2025
48 49
Land Use Regulations
(see Attide 8 for land use
definW:ionJ).
Table 2-13
Office and Industrial Zones
L~D HO-RD U HI
• Data centers are permitted
through the approval of a
Conditional Use Permit ( CUP) in
the Office/R&D and Industrial
zoning districts. Retai l, Service. and Office Uses
Ambulance SOMce• p p
Bank$ and Financial p p Establlslvn6nts. General
Bar«s and Financial
• Ancillary data centers are
permitted by right in the
Office/R&D and Light Industrial
zoning districts.
E&lablishments. Slanckllone p p
ATM
Siainms SHtl<>OI! Conll>rn
can Cen1crs
Oala Contul5
OaLl Cenw,._ Andllarv
Data Center Study Session
Entitlement Process
Updated Zoning Code (2024)
Conditional Use Permit (building use)
Completeness
Applicati0n .i.. Review /Project •
Submittal .,. Clearance .,.
Conunitt~e
Schedule
Planning .i..
Commission .,.
Hearing
p p p p
p p CUP .
CUP CUP CUP CUP
p p p
48 '/' :-"·;, City of
(~ ··J Santa Clara
-c.,..,.'~ Tt . (_-,_- , 1,,ul','il1>1 • t ... ol,I,
Planning
Commission
Decision
Appeal?
lt City
Council
Appeal?
City
Council
Director or
Designee
Decision
Schedule
Development
Review
Hearing
Architectural Review Permit (building design)
Campleteness
_.. Review/Project _... Application
..,.. Clearance ..,.. Submittal
Committee
Item# 2
24 5/20/2025
so
Entitlement Process
Other Jurisdictions - South Bay
• San Jose: Conditional Use Permit in Commercial zones, Special Use Permit in Industrial zones, both involving public hearings • Mountain View: by-right in most Industrial zones
• Milpitas: by-right in most Industrial zones
• Sunnyvale and Cupertino do not enumerate data centers as a use 50 Data Center Study Session ·Y>). City of
( ~ ··-) Santa Clara
~~ 1L._t: .. ,:., ,ll">h•• ,\ •• ,1,1,
51 Entitlement Process
Other Jurisdictions - Virginia
Loudoun County
• Special Exception Permit in Commercial Center, Commercial Light Industry and all Office and Industrial Zoning Districts. - Discretionary review process with Planning Commission recommendation to the Board of Supervisors.
Prince William County
• To promote data center development a "Data Center Opportunity Zone Overlay District" was created where data centers are permitted by right and must meet established design standards. 51 Item# 2
25 5/20/2025
52 53
Nico Procos, Interim SVP Director
Kathleen Hughes, Assistant Director
CA Publicly Owned Utilities Sales Grow
Ranking by 2023 Sales
California Publicly Owned Utilities (>25,000 Customers) Sales Growth from 2019 to 2023
,-- -------- - - --------- - ---- ------ ---- -----, 1 - LADWP 2-SMUD
3- Santa Clara
4 - Imperial Irrigation District
5 - Modesto Irrigation District
6 -Turlock Irrigation District
7-Anshelm
B - Riverside
9 - Roseville
10 - Pasadena
11 - Burbank
12-G lendale
13 - Palo Alto
14- Redding
15 - lod l
16 - Alameda
25%
- 3%
4%
-3% -
- 2%
5%
-7% ______ _
-----•6%
- 2%
-10% -8% -6% -4% -2% 0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% Sales (MWh) Growth from 2019 to 2023
Item# 2
53 26
5/20/2025
54 55
Data Center Growth
60% Power Use and Sales
, 0 are from Data Cent e rs $
Data Ccnt~r revenue
inc:rea.~ d 2 0 fro r,n
FY:10 2.3
.. ~ .000
Data Ccntcl"S usage
increased 4 % from
FY2023 '"'""
2000 00
SVP Load Growth 2010-2025
Customer Expansion - Data Centers, new bulldlnga, or
products 1111,01 0 , ,1no 11 ll lr..-.J12 ,11 ,,0 13 1, , ,-~ 14 11 113;1~ 111/10 16 111/'2017 1110019 ,, 1no 19 1,1 no io 1/ lr."02 1 111;:.,022 111/Xl2 l l/ 1,":'ll24 1/ l !Xl'.:'S 54 - Aetull l Load (MWhs ) - 12•Month Average
Growth & Expansion
• Growth is driven by data centers - request to add
500+ MW load (growing from 700 MW)
• Staff determined there was a growth need in late
2019 - SVP Strategic Planning Session
• Prioritized Growth Plan as an SVP Strategic Priority
• Needs for both SVP Local System and State
Transmission System
• What changed? - The Size
• Most new data centers are 50MW-99MW
• SVP had to tell customers "Limited Capacity"
- 2021 No longer provide will serve letters
- 2022 System Limitations Letter
- 2023 System Impact Studies
55 Item# 2
27 5/20/2025
56 57
Electric Utility Capital Projects
CAPITAL PROJECT 2023 2024 2025 2026 2027 2028 I 2029
I I
CAISO Transmission Project I I I I I Northern Receiving Station (NRS) I I r I I
Kifer Receiving Station (KRS) I I I I
Scott Receiving Station I I I
NRS/KRS 11 SkV Transmission r- I ~
60kV Transmission Upgrades l
Freedom Circle Substation '
CAPITAL PROJE CT DRIVER
San Tomas Substation -
SYSTEM EXPANSION
' Memorex Substation -
DATA CENTER &
LARGE DEVELOPMENTS
Stender Way Substation
-
RELIABILITY
Bowers Avenue Substation I
Laurelwood Substation
Esperanca Substation I
Juliette Substation Expansion I
Kifer 50MW Battery Storage
Homestead Substation I I I
Major Expansion Projects
• Bond Financing
- First Bond Issuance of $130 Million in 2024 and Second Bond Issuance of $240 Million in 2026
• Scott and Kifer Receiving Stations - $229 million
- Full rebuild and expansion
- Replaces end-of-life infrastructure and increases capacity • Schedule
- Bids received on May 1, 2025
-Apparent low bidder is ~6% under engineer's estimate
- Major Material Procurement (In Progress)
- Construction Council Award anticipated in June 2025
Scott Receiving Station (SRS)
Kifer Roceiving Station (KRS)
Item# 2
56 57
28 5/20/2025
58 59
Major Expansion Projects
• Northern Receiving Station - $147 million
- Upgrades and Expansion
- Schedule
• Bids due May 15
• Major Material Procurement (In-Progress)
• Construction Council Award anticipated in June 2025
• 115 kV Transmission line- $48 million
- Connects Northern Receiving Station and Kifer
Receiving Station
- Schedule
• 60% Design Underway
• Easement acquisition - thru May 2026
• Anticipated construction start - late 2026
• Funded through Load Development Fees
(new customers)
• Rate Payers (existing customers)
• Combination of both
• Funding split depends on whether the
project is adding capacity or replacing
infrastructure
Item# 2
58 29
5/20/2025
60 How Power Moves
• 28 Substations & 4 System
Connection Stations
• 33.17 miles of 6okV Power
Lines
Item# 2
574.9 miles of 12kV Distribution
Lines (67% underground)
• 160 Distribution Feeders
• 11,344Power Poles
• 5,000+ transformers
60 Capital Project Funding - An Example
61 • For Discussion Only - Scott Receiving Station
• Replacement of end-of-life equipment and expansion
to accommodate growth
Project Construction estimated at $110 million
• 35% Load Development Fee and 65% Rate Payers
• $38.5 million from LDF
• $71.5 million from rate payers (existing
customers)
• $65.8 million from Industrial (1,288
accounts)
• $4.3 million from Residential (52,529
accounts)
61 30
5/20/2025
Summary of Sales for FY
Category
Residential
Commercial
Industrial
Other
Total
' ■
' . . . '.. . Revenue
(millions)
-
254,000
108,000
5.6%
2.4%
4,181,000 91.6%
17,000 0.4%
4,560,000 100.0%
$40
$24
$6.08
$2
1674
5.9%
3.5%
90.2%
0.4%
100.0%
Residential
90.2 %
Commercial
3.5%
Customer Mix • Revenue 62
62 63
Residential/Industrial Utility Regiona
Rate Comparison
Data Year Utrlrty Name
2023 LAOWP $0.2299
2023 PG&E $0.3404
2023 City of Palo Alto $0.2021
2023 City of Roseville $0.1781
2023 SMUO $0.1689
2023 SVP $0.1475
Industrial Resrdentral
$0.2016
$0.2764
$0.1543
$0.1190
$0.1260
$0.1343
hrgher than
Industrial
14%
23%
31%
50%
34%
10% 63
Item# 2
31 5/20/2025
64 65
SVP Residential - Small Commerc1a
Industrial Rate Comparison
Class of Service SVP Average Rates
efiecltve 01101/25 ISn<Wh '
Residential
Small Commercial
Large Industrial
Very Large Industrial
D-1
$0.175
C-1
$0.259
CB-3
$0.172
CB-6
$0.153
•For genera l reference, average rates based on estimated forecasts, Including surch arges, Individual custome r's averag e rate will depend on It's applicable kw and kwh
•
~Si~~
Pewrll<
~-------
ELECTRIC
RESOURCE
MAP
IN TOWN RESOURCES 95arrtaCl atill.CA
LEGEND •
• Proposition 26 specifies that a "charge
imposed for a specific government service or
product provided directly to a payor that is
not provided to those not charges, and which
does not exceed the reasonable costs to the
local government of providing the service or
product" Article XIIIC of the California
Constitution
• The courts have interpreted this to mean that
no one rate class may subsidize another rate
class.
w ,nd Solar C~prm,JI tt,ct-celearic l.ar.dftll Natura l W S Item# 2
$5
32 5/20/2025
Renewable Energy Procuremen
as a % of. Forecasted
8,000,000
7,000,000
6,000,000
5,000,000
7,075,050 7,200,365 7,337,847 7,409 ,981
66 67
.l!
:S:: 4,000 ,000
::i;;
3,000 ,000
2,000,000
1,000,000
4,867,774 5,133,762
71%
58%
2025 2026
70% 65%
2027 2028
58% 54% 62% 50% · 46% 44%
2029 2030 2031 2032 2033 2034
Under Contract {MWh) In Negotiations {MWh) - •n Pipeline {MWh) -Forecasted Load {MWh) Gary Welling, Director
Ahmed Aly, Principal Engineer
43%
2035
Item# 2
66 33
5/20/2025
68 69
Data Centers & Water
• Data Centers: high water use historically;
however, two major trends influence change:
- Increased adherence to recycled water
use requirement
Adoption of low water use cooling
systems
Recycled Water B
I
Regional Wastewater Facility (RWF)
South Bay Water Recycling (SBWR)
Valley Water Silicon Valley Advanced
Water Purification Center, opened 2014
69 - Current production: blend of treated
effluent yielding improved water quality
- Offset 20% of total water use
Item# 2
~ fRAllCIIG0""1
34 5/20/2025
70 71
Recycled Water Mandate
• Municipal Code 13.15: Mandates maximizing Recycled Water • Rules & Regulations Mandates for Development: Recycled Water Maximize use
Convert existing when allowed by regulations
Extend connections to recycled water system
Recycled Water Ready if not currently connected
Construction use
Implementation of Mandate
• New Data Center Development
Condition use of recycled water
on entitlements and permits
Condition conversion of existing uses to recycled water Condition extension & connection to recycled water system Condition Recycled Water Ready if not currently connected Item# 2
35 5/20/2025
72 73
Current Water Use
Data Centers Using
Recycled Water
Existing
Proposed *6
* All new data centers will maximize
recycled water use
Current Water Use
CITY OF
SUNNYVALE
Data Centers with Recycled Water
• Efficient water demand for future growth
Increased recycled water use
Efficient cooling design & technology
Item# 2
//--:'': '.' City of
( ,_,,,, · ) Santa Clara
- ----~ -~/ Tl ,-f.· -·~ , -,I ',',h.1 • 1',.·.,,t,J.
36 5/20/2025
74 75
Reena Brilliot, Director of Economic Development & Sustainability Regional As~et to Innovation Ecosystem
• Market advantage for companies to site data centers in Silicon Valley:
- Data Centers provide essential service for the Bay Area and State - Region has a high density of technology, social media, and communications companies that are hyper users of cloud services, and thus customers of data storage services.
- Region has strong fiber infrastructure, which is critical for data center operations.
- Need for Data Centers/ AI Factories is growing, technology is being developed in Silicon Valley
Item# 2
75 37
5/20/2025 Item# 2
76 77
General Plan's Fiscal Health Goal
• Data Centers align with City's General Plan Goal 4.6 Maintain the City's Fiscal Health and Quality Services.
• This strategy acknowledges the need for the City to ensure that new growth strengthens and diversifies the City's tax base.
• Data centers contribute funds for services that exceed services the City provides and create a net positive contribution to the City's fiscal health.
Economic Contributions to City
• Data centers contribute annually to
76 City's General Fund through: -~-- -------=-----
- Silicon Valley Power's Transfer
- Property Tax
- Sales Tax
- Business License Tax
• Data centers contribute to Affordable
Housing per City's Affordable
Housing Ordinance 77
38 5/20/2025 Item# 2
78 79
SVP's Transfer
• 5% of SVP revenues are provided to the City's General Fund. • This transfer has been in place since 1951.
• The revenues from this transfer are unrestricted and are used to support City services such as police, fire, street repairs, sidewalks, parks, libraries and senior services.
• November 2022, Voters approved Measure G and authorized the continued transfer of 5% of utility tax revenues to the City's General Fund. • It is estimated that approximately $23.3M in revenue will be provided to the General Fund in FY24/25 from data centers.
78 Property Tax
• Property taxes consist of secured and unsecured assets. - Secured assets are walls, ceilings, floors, etc. of a building. - Unsecured assets are improvements and/or equipment installed within a building.
• Data centers as a property class are unique as they have high value unsecured assets with the number of expensive servers and infrastructure installed.
79 39
5/20/2025
80 81
Property Tax (continued)
• Property taxes contributed by data
centers fund County and City services
and support School Districts.
• Upon construction and operation,
recently developed individual data
centers can contribute typically up to
$200,000-$400,000 in property taxes
to the City annually.
• Data centers are contributing $6.5M
combined in property tax.
Sales and Use Tax
• Sales and use tax contributed by data
centers fund State, City, County and
County Transportation/VT A.
• Data centers are contributing annually
$4.96 M to the City in combined in
sales and use tax.
Who shares property tax
Per $100 Collected
City of
Santa Clara
$10.17
r---
County
Office of
Education
$3.97
Other Special
Districts
$2.51 ERAF
(Other School
/
Districts)
$15.19
Santa Clara
Unified School District
$38.36
Who shares sales & use tax
Per $9.125 Collected on a $100 Purchase
County
Transportation/
VTA/Caltraln.
including
MeasureB ~
and BART
$2.00
5anta Clara County -
General Purpose
$0.125
Item# 2
80 81
40 5/20/2025
82 83
Affordable Housing Contributions
• Since the Affordable Housing Ordinance's
effective date in 2019, data centers have
paid $3.9 million toward affordable
housing.
• Three data centers in the pipeline, are
anticipated to provide $1.75 million for
affordable housing.
Estimated Total 24/25 FY General Fund
Contributions
• $29.5M - Estimated SVP's Power Payment In-Lieu of Tax • $6.5M - Property Tax
• $4.96M - Sales and Use Tax
82 • $ 40.96M Estimated Total Contribution to General Fund (combined SVP Transfer+ Property Tax+ Sales and Use Tax)
• Total Contribution is Estimated to be 13% of General Fund B3
Item# 2
41 5/20/2025 Item# 2
White Paper
Overview of Industry
Data Center Markets & Types
Building Security & System Security
Energy Sources & Impacts
Economic Impacts
Challenges
Summary
84 Data Center Study Session ;;-<t .: City of
1, ·~ ' ) Santa Clara
. -...:..._•-.......:.;_,,,- 11,,: l· ~ ,,,,, ,,I ',','I,, ,, ; , • . ,\,I, Summary
Background
Industry Insights
Data Center Land Use Regulations & Aesthetics
SVP's Power Supply and Power Planning for Data Centers
Data Centers and Water
Economics of Data Centers
Questions
85 85
42 5/20/2025
86 City of
Santa Clara
The Center of What's Possible
Item# 2
43 From:
To:
Masheika Allgood
Qfilk
Subject:
Date:
Public Comment - May 20, 2025 Hybrid Meeting - Agenda Item 25-448 Tuesday, May 20, 2025 5:01:39 PM
Attachments: There isn"t enough water for all of us Qdf Who needs air anyway.pdf
I You don't often get email from [email protected]. Learn why this 1s important My name is Masheika Allgood and I am an AI ethicist. My work is largely focused on the environmental impacts of AI data centers. I'd like to include the two attached articles in the meeting minutes as reference materials on the water requirements for AI data centers and the air pollution from co-locating gas generators.
Masheika Allgood
Founder
AIIAI Consulting, LLC I allai-us.com
Environmental Platform: Taps Run Dry
[email protected]
POST MEETING MATERIAL
5/20/25, 4:59 PM There isn't enough water for all of us There isn't enough water for all of us
On the importance of fairly negotiating data center contracts Q9
MASHEIKA ALLGOOD
APR 02, 2025
D Share
I've been writing about AI data center-driven water scarcity for some time. The mo common response to my work is - "water is never lost to the earth." Which is True. we don't live on the earth. We live in communities. So while the earth may have enough water to support all of humanity, your community may not have enough wa to support the people living in it. Just as we grow enough food on the planet but wb societies are suffering famine, the existence of water on the earth does not protect your community from water scarcity. Water scarcity is driving migration into cities further exacerbating the pressure on munici pal water infrastructure .
So, how do we protect our water? Or at least, slow down AI companies' gluttony un they build technologies that can sustainably address their water needs at scale? To start, we need to put municipal leaders in a position to fully negotiate data center contracts.
All contracts are negotiable
I had a conversation with a friend recently about a horribly onerous contract she signed for a speaker agent. The contract either required up front payment of all fee:
it charged a higher fee plus interest for monthly payments. This wasn't a lease agreement, credit card, or mortgage, it was for someone to help get her talks . And,, course, after my friend paid in full, they got shoddy service (ignored, talked over, ge generated speaking applications, etc.). One of the first things I learned in running n https ://eatyourfrog .substack .com/p/there-isnt-enough-water-for-all-of 1/11 5/20/25, 4:59 PM There isn't enough water for all of us own real estate/civil litigation firm was - never sign the other party's contract. It always sucks.
Intentionally so. We're taught in law school to create contracts that are completely one-sided towards our client, with the understanding that opposing counsel will do the same. The idea is that you both start at opposite ends and negotiate your way to some sort of middle ground. Which is why you will always lose if you sign someone else's contract. You have to bring a counterbalance to the table so you can start negotiating your way to a middle ground.
So what does all of this have to do with data center water consumption? Well, everything. Water is a municipal concern. It's managed at the local level by cities, counties, and special water districts. But I have yet to meet a local official who had understanding of how much water the data center they approved would require to operate, or how much it would ultimately consume.
We've been signing tech company contracts, with no counterbalance . And they suck.
So how did we get into this situation, and how do we remedy it? Math. In order to negotiate, our municipal leaders need to have some understanding of how much wa the data center will require before approving the build - during the permitting and zoning process. Which is hard. Well, it was hard.
State of water loss calculations
If you search for how to calculate cooling tower water loss, you'll get a host of articl which all reference the Makeup Water calculation:
https ://eatyourfrog.substack.com/p/there-isnt-enough-water-for-all-of 2/11 5/20/25, 4:59 PM There isn't enough water for all of us Makeup water= Drift loss+ Evaporation loss+ Slowdown loss Makeup water is the water added back to the system to compensate for losses. Thin of it like tire pressure. As you ride your bike, or drive your car, air leaves the tire for variety of reasons (heat, pressure, holes, etc.). If the tire loses enough air it can suffc catastrophic failure, so you want to refill the tire to its optimum level regularly. Thci the principal behind makeup water. The cooling tower, and all of the systems that r on it, risk suffering catastrophic failure if the water level gets too low. So you want refill the water to its optimum level regularly.
The makeup water calculation considers all the ways cooling towers lose water. Wh you add all of those losses up, you know how much water you have to put back into system to keep it at the optimum level. While the base calculation doesn't look too daunting, the complexity is in the loss calculations.
Drift loss is the water that the wind blows away. It's heavily impacted by the shape the cooling tower (some let in more wind than others). But wind is too variable and unpredictable for a granular calculation. So drift loss is usually calculated as a percentage of the water that's being circulated through the cooling tower.
The complexity ramps up when calculating evaporation and blowdown loss.
Evaporation loss
Evaporation loss is the water that is evaporated during the heat transfer process (se, The Fallacx. of_ Closed Loop_ Cooling Sx.stems for more details). Here's the evaporation lo~ calculation:
E= CX(Ti-To)XCJJ
.,\
Image pulled from Hm Do You Calculate Water Lo in a Coo.ling Towe r https://eatyou rfrog .su bstack. com/pith ere-isnt-en ough-water-for-all-of 3/11 5/20/25, 4:59 PM There isn't enough water for all of us This calculation requires you to know:
• The rate of water circulation (C)
• The difference in water temperature from the bottom of the tower to the top of the tower (Ti-To)
• And how much heat was added to the water during the heat transfer process (C None of which is available when a data center is being planned.
Eat Your Frog is a reader-supported publication.
Blowdown loss
To receive new posts and support my work,
consider becoming a free or paid subscriber.
The blowdown loss calculation isn't particularly complicated, but it also relies on specific systems operation data. Blowdown loss is the water that is removed from tl system because it's become to concentrated with sediments. Think of it as the dreg:
what's left at the bottom of a coffee pot or a bottle of (unswirled) red wine. The sediments are super concentrated in that last glass. The same happens in engines. I why regular oil changes are a key part of car maintenance. As the oil cycles through the engine some evaporates due to heat, and it picks up contaminates from the engi Over time, the oil gets thicker and starts to concentrate into a sludge. Sludge doesn lubricate the engine well, so your car starts to perform poorly.
Blowdown is water that is heavily concentrated with solids that were picked up as i cycled through the cooling system. Data center cooling systems are more sensitive than car engines so, instead of waiting for the water to form a sludge and changing all at once, data center operators expel some amount of blowdown regularly so the system is never impacted.
https ://eatyourfrog .substack.com/p/there-isnt-enough-water-for-all-of 4/11 5/20/25, 4:59 PM There isn't enough water for all of us Your company comes with a guide for when you should get oil changes based on th amount of miles you drive. Those miles are a proxy for the amount of times the oil l cycled through the engine. The blowdown calculation is based on the same principl The core calculation is the cycles of concentration (CoC):
Blowdown = [Evaporative loss - (COC - 1) x Drift loss] /(COC - 1) To calculate blowdown, in addition to evaporative loss and drift loss, you need to kt the cycle of concentration for the cooling tower. Which is where things get trick.
There is no proxy for how many times the water cycles through the system. Instead the cycle of concentration is calculated one of two ways
1. By the ratio of chloride content in the circulation water v makeup water
2. By the ratio of conductivity of the system water v makeup water Neither ratio is derivable when a data center is being planned.
While the makeup calculation looks straightforward, it is technical and confusing ii practice. It relies on specific systems operation data. Data that isn't available durin1 the data center planning process and is closely held as proprietary information one, the data center is in operation. Data center operators are the only people who have access to the data required for these calculations and they currently have no legal obligation to report any of this information at any time.
So what do we do when we're outside looking in? How do we obtain the informatio necessary to level the playing field in negotiations? As I mentioned in this IEEE/OECD session during the France AI Summit - we move forward by recognizi1 the audience and the purpose of the calculations. Municipal leaders don't need scientifically accurate conclusions, they need directional forecasts. Thankfully, ther math for that!
https://eatyourfrog .substack .com/p/there-isnt-enough-water-for-all-of 5/11 5/20/25, 4:59 PM There isn't enough water for all of us The hidden costs of Al: Unpacking its energy and water footprint a
Energy-based calculation
The one piece of data we always have about a data center at the earliest planning phases is the electricity it will require. Which is precisely what Uptime Institute utilizes as the key input when for its cooling tower makeup calculation. Uptime Institute created the ToP-ology: Tier Standard as "a performance benchmarking syst to help data center owners and operators identify the performance capability of the data center infrastructure." Data centers are ranked in tiers based on their performance on the Standard's benchmarks. Data centers can be certified at the de~ level , or while in operation. The makeuP- calculation for design-level certification utilizes the following assumptions:
• Each 1,000 kilowatts of cooling load requires 1,027 US gallons per minute of condenser water flow through the evaporative cooling towers o Cooling system utilizes 3 imperial gallons per minute of condenser water p ton of cooling
https://eatyourfrog .substack.com/p/there-isnt-enough-water-for-all-of 6/11 5/20/25, 4:59 PM There isn't enough water for all of us • Evaporation consumes about 1 % of condenser water flow • Drift and blowdown consume about .5% of condenser water flow By tying water flow to cooling load, Uptime Institute made it possible to calculate makeup water based solely on the amount of energy used for cooling:
Assumptions :
a. Each 1,000 kilowatts (kW) of cooling load (approximately 285 tons refrigeration [TAJ) requires about 855 Imperial gallons per minute• (gpm) of condenser water flow (1,027 U.S. gpm or 3,887 liters per minute) throu£ the evaporative cooling towers. at 3 gpm of condenser water per ton of cooling. b. For the purposes of estimating water requirements, evaporation consumes about 1 % of condenser water fie drift and blow-down consume another 0.5%. Thus, a source of water is required to replenish-or "makeup" -a
1.5% of condenser water flow to sustain evaporative cooling processes.
Using the above assumptions, the amount of makeup water necessary to sustain evaporative cooling for a 1,000-kW load for 24 hours is:
(855 gpm) x (60 minutes/hour) x (24 hours/day) x (1.5%) = ::: 18,500 gallons (22,218 U.S. gallons or 84,103 liters).
The only variable in this calculation is cooling load. The current assessment of the amount of a data center's energy that is used for cooling is 40%. With all of the assumptions accounted for, we can do the math!
Let's use the Stargate AI data center campus in Abilene, Texas as an example. The c center is intended to run on 360 megawatts of P-OWer. How much water would a dat:
center that runs on that amount of power use and consume? 360MW x 40% = 144MW cooling power
divide by 1666 KW= 144KW
144KW x 1027gpm = 147K gallons used per minute
147k gallons x 60 = 8.8M gallons used per hour
https://eatyourfrog .substack .com/p/there-isnt-enough-water-for-all-of 7/11 5/20/25, 4:59 PM There isn't enough water for all of us
8.8M gallons x 24 = 212M gallons used per day
212M gallons x 1.5% = 3.2M gallons makeup water per day
3.2M gallons x 30 = 95M gallons makeup water per month
95M x 12 = 1.1B gallons makeup water per year
Based on these calculations we can estimate that the Abilene, Texas Stargate AI dat center campus will require 212 million US gallons per day to cool. Of that 212 millic US gallons, 3.2 million gallons will be consumed. Consumed means lost to the community it was taken from. The evaporated and drift water will move through th global water cy:cle and the blowdown will need to be treated . There are systems to minimize drift and reuse some percentage of blowdown, but they are expensive. Yo1 can't assume companies will implement them if they aren't required to. But that's t1 point of this exercise, to empower leaders to start those mitigation discussions as e; in the process as possible.
DIV Water Loss Calculator
Now that we have a publicly accessible calculation - let's make it public!! AllAI Consulting, Inc. utilized the Uptime Institute's Tier Standard: Topology makeup w, calculation to create the world's first public Data Center Water Consumption Calculator (available here ).
https ://eatyourfrog .substack .com/p/there-isnt-enough-water-for-all-of 8/11 5/20125, 4:59 PM There isn't enough water for all of us Total Data Center Electricity
[ 360.00MW
Select Unit
Gallons Liters
Cooling Powet
Generated from the Data C nrer Water Con umP.tion Cal ulacor The estimates provided here can serve as the starting point for discussions on data center water use and consumption. Understand that the amount of water that is requested from your aquifer is not the sum total of the water required to cool the d, center. Most of the water will not come from your city's aquifer, but it's coming fror somewhere. And given the cost and complexity of piping/shipping water from othei states and regions, it likely isn't coming from too far away. Having a sense of the overall system requirements places you in a position to assess the validity of an operator's water resourcing plans. Having an initial estimate enables you to ask mo consequential questions early in the process. Specifically:
\) How do the assumptions in your estimate differ from those in the tool? https :/leatyourfrog .substack.com/p/there-isnt-enough-water-for-all-of 9/11 5/20/25, 4:59 PM There isn't enough water for all of us Q What is your average facility-level water consumption percentage? 9 On average, how many hours in a year do your facilities run at peak usage? 9 What is your local water supply chain and what percentage of the estimated usaf can it meet today?
9 How much of the estimated consumption do your replenishment efforts account for, and in what timeframe?
While all contracts are negotiable, drinkable water is finite. There isn't enough wat for all of us and all of the AI data centers companies want to build. Not given the current technology. But there doesn't have to be. Municipal leaders are charged wit using zoning and permitting processes to ensure there's enough water for the population first. Now that this tool provides communities a legitimate seat at the negotiating table, it's critical that we secure our survival first. It is a company's job secure the resources they need to be successful, not humanity's. If they cannot oper their businesses within the reality of the drinkable water that exists on the earth - t perhaps they should innovate. Faster.
Eat Your Frog is a reader-supported publication.
To receive new posts and support my work,
consider becoming a free or paid subscriber.
9 Likes · 3 Restacks
Discussion about this post
Comments Restacks
https://eatyourfrog .substack.corn/p/there-isnt-enough-water-for-all-of 10/11 5/20/25, 4:59 PM There isn't enough water for all of us Write a comment...
© 2025 Masheika Allgood · Privacy · Terms· Colle ion notice Substack is the home for great culture
https://eatyourfrog .substack.com/p/there-isnt-enough-water-for-all-of 11/11 5/19/25, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog Who needs air anyway?
Data center and generator co-location as a threat to human survival. 01 MASHEIKA ALLGOOD
MAY 02, 2025
0 a Share
I shifted my AI ethics advocacy to a focus on water because my key concern has al"\\ been community survival and human thriving. Communities cannot exist without drinkable water. They also cannot survive without clean air. https://eatyourfrog.substack.com/p/who-needs-air-anyway 1(7 5/19/25, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog ~
C
co a::
25 Cities Most Polluted by Daily PM
1 2
3 Bakersfield, CA
Fresno-Madera-Hanford,
4 Eugene-Springfield, OR
5 6 Reno-Carson City-Fernley, NV
7 8
9 San Jose-San Francisco-Oakland, CA
Redding-Red Bluff, CA
Sacramento-Rosevtlle· cp.
10 11
12 13
14 Chico. CA
Los Angeles-Long Beach, CA
Yakima. WA
Seattle-Tacoma. WA
Missoula. MT
15 Spokane-Spokane Valley-Coeur d'Alene, WA-ID
16 Phoenix-Mesa. AZ
17 Boise City-Mountain Home-Ontario, ID-OR
18 - San Diego-Chula Vista-Carlsbad, CA
19 - Salt Lake City-Provo-Orem. UT
20 - Denver-Aurora. CO
21 - Logan, UT-ID
21 - Portland-Vancouver-Salem, OR-WA
23 - Las Vegas-Henderson. NV
24 - Fargo-Wahpeton, ND-MN
24 - Salinas, CA
0.0 10.0 20.0 30.0 40.0 50.0
Weighted Average Days
American Lung Association cace of che Air 2024
60.0 70.0
I live in the Bay Area of California, the epicenter of the AI boom. According to the American Lung Association, the Bay Area is the 7th most P-Olluted city: in the count in terms of fine P-article P-ollution . The Bay Area is the seat of Silicon Valley and ho https://eatyourfrog.substack.com/p/who-needs-air-anyway 2{7 5/19/25, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog to 140 of Cal ifornia,s 311 data centers . There are several additional data center proj1 in the works, including a recently approved 99MW Microsoft data center in Mount, View, and the fast-tracking of a 99MW data center housin g comP-lex in San Jose. Bu what is the connection here? Why am I talking about data centers in an article abot air pollution?
The answer is simple - generators.
Generators are essentially a backup technology - a failsafe to enable data center operators to keep their systems running in the case of a P-OWer failure . Or, at least, t were. While most of the articles you'll find on data center generators discuss them i this traditional role, AI data centers are fundamentally breaking with that construe AI data centers are so power hungry that our electricity. infrastructure cannot kee P Responsible operators would plan for the constraint and find innovative ways to operate within these defined limitations until more power was made available. But is tech. Responsibility isn't really our thing.
So we're filling the gap with generators. And we're not talking about a gap of hours an occasional basis, data center operators are running gas powered industrial size generators 24/7, with no stated end date. xAI built a 100,000 GPU data center in https://eatyourfrog .substack .com/p/who-needs-air-anyway 3(7 5/19/25, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog Memphis, Tennessee last year. The data center, which is being scaled up to 200,000 GPUs, requires 150MW of power to operate at the 100,000 GPU level. The Memphi power grid can only support 8MW of power. So xAI is using 35 'temP-orary.:' methan , gas burning turbines to power the facility. The generators are running all day, every day, producing 420MW of power, with no decommission plan. xAI isn't an anomaly, in the face of power constraints tech is turning to generators primary source of electricity. The practice of using generators to power data center operations is called "co-locating." It has become such a standard practice that the Federal Energy Regulatory Commission has been reQ_uired to hold P-roceeding,s. to address cost allocations between data center operators and grid operators, along wi grid stability and reliability concerns. These rules are necessary because companies increasingly creating their own infrastructure for these generator-first data centers di rectlY-con nectin g..!.Q._gas P-lants. PiP-eline OP-erators are overwhelmed by. re~uests from data center operators to expand their pipeline networks to accommodate dire, data center connections. Co-location is quickly becoming the norm, and data cente operators are single-handedly revitalizing the natural gas industry..
Eat Your Frog is a reader-supported publication.
To receive new posts and support my work,
consider becoming a free or paid subscriber.
But this isn't a climate change discussion, it's about human thriving. So why am I talking about natural gas?
According to the American Lung Association, the Bay Area is the 7th most pollute, city: in the country. in term of fine particle pollution . The Bay Area is the seat of Silicon Valley and home to to 140 of California s 311 data centers . Data centers that increasingly powering their operations using gas-powered generators. You know w kind of pollution gas-powered generators produce? Fine P-article pollution . https://eatyourfrog .substack .com/p/who-needs-air-anyway 4{7 5/19/25, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog Fine particulate matter pollution, also known as PM2_5, consists of particles measur
2.5 micrometers or smaller in diameter. These particles are uniquely capable of penetrating deep into lungs, making the health impacts of exposure very serious. E1 short-term exposure to PM2_5 (hours to days), has a significant imP-act on mortalitY rates in urb~n areas . xAI's generators have been running 24/7 for months.
We cannot afford the world we're building. The world doesn't have enough clean ai:
drinkable water to support both humanity and the technodream the zealots among~ us are attempting to impose on all of society. We keep waiting for these folks to eitl find their 'better angels' or begin to recognize the value of long-term strategic development but they are fervently and adamantly opposed to the eating of frogs. T have a vision for our future that doesn't require any of us to have clean air as a generality. It just needs to be clean in our workspace. And our home. Which is ... ou1 workspace ...
San Jose just fast-tracked an application to build a data center and housing comP-le~ the middle of downtown . Data center and housing. Why wait for dispersion of the a P-ollution from all of the data centers located throughout the Bay Area to kill us slm when we can just plop a data center in the middle of a downtown housing complex suck the pollution directly from the source?
https ://eatyourfrog.substack.comlp/who-needs-air-anyway 5r, 5/19/25, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog So, yeah, I'm talking air pollution now. Because, life. Life matters. And I'd like to thank Roishetta Sibley Ozane and Moms Clean Air Force for showing me the way.
Our communities, your community cannot survive without clean air and drinkable water. And they are both at risk in this data center ramp~P- - Abilene, Texas , they're coming for you first but the list of affected states is lon g. It's time to enter the fight Join us at TaP-s Run Dry, there is work that only you can do. We're here to help you 1 it.
Q 1 Like
Discussion about this post
Comments Restacks
https ://eatyourtrog .substack .com/p/who-needs-air-anyway 6(7 5/19125, 10:41 PM Who needs air anyway? - by Masheika Allgood - Eat Your Frog Write a c~mment...
© 2025 Masheika Allgood · Privacy· Terms· Collection notice Substack is the home for great culture
https:1/eatyourfrog.substack.comlp/who-needs-air-anyway 7f7 From:
To:
Cc:
Subject:
Date:
Attachments:
Mayor and Council
Mayor and Council
Afshan Hamid; C!.ru;
FW: Data Center Discussion
Tuesday, May 20, 2025 5:27:23 PM
imaaeoo1.ona
Dear Mayor and Council,
We received the following email which we are forwarding for your reference .
Thank You,
Melissa Lee I Executive Assistant
Mayor & Council Offices I City of Santa Clara
(408) 615-2252 I www.santac laraca.gov
City of
Santa Clara
The Centcr of Whafs Possible
From: Tahir Nairn <tahirjnaim@yahoo .com>
Sent: Tuesday, May 20, 2025 4:00 PM
To: Mayor and Council <[email protected]> Subject: Data Center Discussion
Hello,
I understand data center developments will be discussed tonight. I hope part of that discussion will focus on SVP commercial rate differentials vs PG&E. They seem to be
40% less and I wonder if we should have such a
differential. Maybe only 20% less?
Should there be a tier or separate set of rate tiers for data centers?
I imagine water use by data centers may be separately
regulated, but siting in our fair city may want to consider ,POST MEETING MATERIAL
impact on our groundwater. I believe where I live on
Graham Lane relies on local wells rather than Hetch
Hetchy.
Should there be design requirements for passive cooling via the building design?
Can we "wrap" other use-types around these buildings
such as apartments, offices or hidden parking?
I believe I wrote earlier about the use of data center
waste heat in district heating systems. The data centers likely cannot make best use of this heat for powering
cooling, but our industrial areas may be able to be
redeveloped to have district distribution in mind.
Sincerely,
Tahir J. Nairn
i
0 Approved Data Center Locations ,n Santa Clara
!
~
~ BUNKER HllllN
Tfi,SMANO~
""10Cl>ACYWY
i!i :,j 1<
,. #Q .. !!
15 i! ;f
..
\
~
, \
! . ;_
I
-l~
,:
d
8 '~
~
lls10,
"'°""'•r
•
SCorr.,_VD
-i ••• ....
KIF<RRD
1;-
WAlSHAVf.
• ...
I
MARTINA'lf
MONROE ST
I
0.5 Miles
CAIRJUOAVf
Approved I Built / Under Construction:
1. 5101 Lafayette St 17. 3005 Coronado Dr 33. 3075 Raymond St 49. 1700 Space Park
2. 4700 Old Ironsides Dr 18. 2820 Northwestern Pkwy 34. 510 Mathew St so. 2175 Martin Ave
3. 4650 Old Ironsides Dr 19. 2880 Northwestern Pkwy 35. 3060 Raymond St 51. 2050 Martin Ave
4. 2807 Mission College Blvd 20. 2840 Northwestern Pkwy 36. 3045 Raymond St 52. 1700 Richard Ave
5. 2805 Mission College Blvd 21. 2915 Stender Way 37. 3011 Lafayette St 53. 1100 Memorex Dr
6. 2305 Mission College Blvd 22. 2901 Coronado Dr 38. 1725 Comstock St 54. 737 Mathew St
7. 2151 Mission College Blvd 23. 2895 Northwestern Pkwy 39. 1525 Comstock St 55. 737 Mathew St
8. 2201 Laurelwood Rd 24. 2625 Walsh Ave 40. 1101 Space Park 56. 1231 Comstock St
9. 3030 Corvin Dr 25. 2600 Walsh Ave 41. 1111 Comstock St 57. 2055 Lafayette St
10. 3000 Corvin Dr 26. 2403 Walsh Ave
~
11. 2220 De La Cruz Blvd 27. 2401 Walsh Ave
12. 2970 Corvin Dr 28. 3105 Alfred St
42. 1201 Comstock St Approved I Unbuilt: 43. 1100 Space Park
44. 1500 Space Park 58. 2805 Bowers Ave
13. 3035 Stender Way 29. 3205 Alfred St
14. 2972 Stender Way 30. 1350 Duane Ave
15. 2950 Stender Way 31. 3080 Raymond St
45. 2805 Lafayette St Pending Approval
46. 1550 Space Park
47. 1160 Walsh Ave 59. 651 Martin Ave
16. 3020 Coronado Dr 32. 3223 Kenneth St 48. 651 Walsh Ave POST MEETING MA~ \ "'
·1: