Data centers & energy: Did w id we ge get it t it backwards - - PowerPoint PPT Presentation

Data centers & energy: Did w

id we ge get it t it backwards ds?

Adam Wierman, Caltech

The typical story about energy & data centers:

The typical story about energy & data centers: Su Sust stainab able d dat ata a centers

Remember: The cloud is (often) more efficient than the alternative.

The typical story about energy & data centers: But maybe we got it backwards?

Renewable energy is coming! …but incorporation into the grid isn’t easy

Key Constraint: Generation = Load

(at all times)

low uncertainty

Today’s grid

Generation Load

Key Constraint: Generation = Load

(at all times)

low uncertainty controllable (via markets)

Today’s grid

Generation Load

Key Constraint: Generation = Load

less controllable high uncertainty

(at all times)

Tomorrow’s grid

low uncertainty

Key Constraint: Generation = Load

less controllable high uncertainty low uncertainty

(at all times)

1) 1) Huge ge pri price v vari riability, leading to generators opting out of markets! 2) 2) More c e conven enti tional res eser erves es n needed eeded, countering sustainability gains!

Key Constraint: Generation = Load

less controllable high uncertainty low uncertainty

(at all times)

1) 1) Huge ge pri price v vari riability, leading to generators opting out of markets! 2) 2) More c e conven enti tional res eser erves es n needed eeded, countering sustainability gains!

Grid needs huge growth in demand response (or storage) Data centers are a promising option

…they are large loads …usage is growing quickly …highly automated …they have significant flexibility 500 k 500 kW-10 100 M 0 MW each 10 10-15% grow growth/year

Data centers are a promising option

…they are large loads …usage is growing quickly …highly automated …they have significant flexibility cost flexibility

10+ years of research into energy-efficient data centers

Build lding m manage geme ment

5% in 2 min / 10% in 20min [LLNL] e.g. cooling, lightin g, …

Data centers are a promising option

…they are large loads …usage is growing quickly …highly automated …they have significant flexibility cost flexibility Build lding m manage geme ment

5% in 2 min / 10% in 20min [LBNL] e.g. cooling, lighting, …

10+ years of research into energy-efficient data centers

Data centers are a promising option

…they are large loads …usage is growing quickly …highly automated …they have significant flexibility cost flexibility Workloa load m manage geme ment

10-30+% in 10-60min [LBNL,HP] e.g. demand shaping, geographical load balancing, quality degradation, …

10+ years of research into energy-efficient data centers

Data centers are a promising option

…they are large loads …usage is growing quickly …highly automated …they have significant flexibility cost flexibility Microgri rogrid manage geme ment

10-100% in 5-30min e.g., Battery management, local PV, Backup generation

10+ years of research into energy-efficient data centers

A new story about energy & data centers: Data centers are valuable resources for making the grid sustainable

=

What t is is th the p pote

• tential of
• f data

ta ce center de demand r respon

• nse?

Optimally placed, fast charging rate storage

interactive workload batch workload PUE

PV

A A case s study tudy:

data center

A A case s study tudy:

A A case s study tudy:

$1 $1-5 m milli llion

• n c

cost! 1 1 MWh Wh if geogra raph phica cal l load b bala lanci cing g is used!

Where a are we we toda today?

Data centers rarely participate … and if they do it is highly inefficient

Time of use pricing Coincident peak pricing Wholesale markets Ancillary service markets Emergency DR

Where a are we we toda today?

• Risky to participate
• Few opportunities for utility

to extract response For more see [Liu et al 2013]

peak coincident peak warnings

Time of use pricing Coincident peak pricing Wholesale markets Ancillary service markets Emergency DR

Where a are we we toda today?

Time of use pricing Coincident peak pricing Wholesale markets Ancillary service markets Emergen ency cy D DR

How ca

• w can we

we do do bett tter?

Algorit ithm desig sign fo for da data c cen enter ter p parti ticipation New ew marke ket des t designs Eng ngine neering ng: Economic ics: s:

+

[Camacho et al 2014], [Chen et al 2013, 2014], [Ghamkhari et al 2012, 2014], [Aikema et al 2012, 2013], [Irwin et al 2011], [Urgaonkar et al 2013, 2014], [Li et al 2012, 2013], [Liu et al 2013, 2014] [Liu et al 2014], [Chen et al 2014],[Ghamkhari et al 2013], [Li et al 2013], [Wang et al 2014], [Ren et al 2015], [Wierman et al 2014], [Zhang et al 2015]

How ca

• w can we

we do do bett tter?

Algorit ithm desig sign fo for da data c cen enter ter p parti ticipation New ew marke ket des t designs Eng ngine neering ng: Economic ics: s:

+

…bu but, a adopt ption re repre prese sents s a huge c challe llenge

Performance is priority #1 Highly regulated, change is difficult

A A sta tarti ting p poin

• int: Col
• loca
• cated (mu

(multi-tenant nt) d ) data cent enter ers

29

Hyp yper er-scale ( (e.g. .g. goog google): 7 7.8 .8% Ent nterprise: 53% 53% Coloc location

• n: 3

: 37% …of total data center industry electricity usage

A A sta tarti ting p poin

• int: Col
• loca
• cated (mu

(multi-tenant nt) d ) data cent enter ers

Why Why col coloca

• cated data cent

enter ers?

…but th t the e da data ta c cen enter ter w would l like to ke to participate i te in dem demand r res esponse! e! Building ope g opera ration i is sepa parated f from rom com computing pri priorities + On-site generation provides backup! + Market power isn’t an issue! + No regulation – can do whatever they want! + Tenants are heterogeneous in workloads! Set u et up incen enti tives es for ten tenants ts

Lot’ Lot’s of

• f wor

work to b k to be don

• ne…

The typical story about energy & data centers: But maybe we got it backwards? Key p y point nts: 1) ) We ne need t to move b beyo yond a a “myo yopic” f focu cus o

• n

n a da data ce centers t to co cons nsider a a “sy syst stem-wi wide de” v view o w of sustaina nability. ty. 2) It is t is im important to c to con

• nsider m

mor

• re th

than g goo

• ogle-like “

e “hyper per-sc scal ale” d dat ata a centers. s.

Data centers & energy: Did w

id we ge get it t it backwards ds?

Adam Wierman, Caltech