Jonathan Koomey: The Environmental Cost Of Cloud Computing

The Environmental Cost of Cloud Computing: Assessing Power Use and Impacts Jonathan G. Koomey, Ph.D. http://www.koomey.com

Lawrence Berkeley National Laboratory & Stanford University Presented at Green:Net San Francisco, CA March 24, 2009 1

For users, the cloud offers infinitely scalable computing on demand

2

So why should cloud users care about power use?

3

Power use strongly affects costs for “in-house” IT services (the alternative to relying on the cloud) AND

4

Cloud computing suppliers have two inherent advantages on power and costs over “in-house” IT (load diversity and economies of scale) 5

(As an aside, most people think the true total cost for “in-house” IT is far lower than it actually is)

6

Data centers, where the cloud resides, are where the world of bits meets the world of atoms

7

8

The cloud uses electricity. How much?

9

World data center electricity use, 2000 and 2005

Source: Koomey 2008 10

How much is 152B kWh? Italy South Africa Mexico World Data Centers Iran Sweden Turkey 0

50

100

150

200

250

300

Final Electricity Consumption (Billion kWh) Source for country data in 2005: International Energy Agency, World Energy Balances (2007 edition) 11

Trends push power use both up and down

12

Pushing power use up… •  Increasing demands for –  E-commerce –  VOIP –  Internet search –  software as a service –  video downloads –  resilience in the face of disaster –  regulatory compliance (e.g. Sarbanes-Oxley) –  IT-enabled business transformation

•  More transistors on a chip + more RAM + more volume servers 13

Summary: Delivery of IT services is increasing rapidly

14

Pushing power use down… •  •  •  • 

Virtualization/consolidation Cooling and power constraints Recognition of constraints by the C level Metrics –  Servers + other IT equipment (Spec Power, 80 plus, E*) –  Site infrastructure

•  Utility rebates (PG&E) 15

Summary: Information technology is becoming more energy efficient at a furious pace

16

Internet electricity intensity Electricity per GB transferred down 30% per year!

Source: Taylor and Koomey (2008) for 2000 and 2006 data. Trends for 2000 to 2006 extrapolated to 2008 by JK.

17

Data center costs are strongly affected by IT power use

18

Annualized data center costs

19

Two important equations

Power related terms

20

In spite of our historical progress, there’s still great potential for improving the energy efficiency of data centers 21

Many efficiency opportunities

Source: EPA report to Congress 2007

22

Improving the energy efficiency of data centers is as much about people and institutions as it is about technology 23

Efficiency opportunities •  Improve asset management and utilization (multiple benefits) •  Improve efficiency of components (e.g. power supplies) •  Implement consistent metrics and track over time •  Align incentives to minimize True Cost of Ownership •  Think “whole system redesign” (RMI) 24

Misplaced incentives •  Energy, efficiency, and performance metrics not standardized •  Not charging per kW but per square foot •  Split accountability –  Who pays the bills, IT or facilities? –  Who bears the risk of failure?

•  Hierarchy and culture differences •  Piling safety factor upon safety factor •  Not focusing on total costs for delivering computing services 25

Cloud computing suppliers have at least two big advantages on power and costs over “in-house” IT

26

1) Diversity: spread loads over many users, improving hardware utilization

27

2) Economies of scale: implementing technical + organizational changes is cheaper and easier than for small IT shops 28

The biggest environmental story about information technology (IT) is not direct electricity use (which is relatively small) but how IT affects efficiency in the broader society 29

Why?

30

Moving electrons is always less environmentally damaging than moving atoms

31

Example: paper vs. PDF •  Mass of paper = 5 g/sheet •  Mass of electrons to move a 1 MB PDF file of that page (based on average network electricity intensity of 7 kWh/ GB) is 1.7 x 10-5 g •  Ratio of paper mass to electron mass ~ 300,000 32

Conclusions •  The cloud is responsible for 1-2% of the world’s electricity use. •  Absolute electricity use growing fast (doubling every 5-8 years) •  IT services are growing much faster than electricity use (doubling every year or two). •  Electricity productivity, defined as computing services delivered per kWh, is increasing rapidly and this trend promises to continue. •  The indirect environmental and productivity benefits of IT are likely to be more important than direct electricity use. 33

Key web sites •  EPA on data centers + 2007 Report to Congress http://www.energystar.gov/datacenters •  LBNL on data centers: http://hightech.lbl.gov/ datacenters.html •  The Green Grid: http://www.thegreengrid.org/ •  The Uptime Institute: http://www.upsite.com/ TUIpages/tuihome.html

•  SPEC power: http://www.spec.org/power_ssj2008/ 34

References • 

Koomey, Jonathan. 2007a. Estimating regional power consumption by servers: A technical note. Oakland, CA: Analytics Press. December 5. (http:// www.amd.com/koomey)

• 

Koomey, Jonathan. 2007b. Estimating total power consumption by servers in the U.S. and the world. Oakland, CA: Analytics Press. February 15. (http:// enterprise.amd.com/us-en/AMD-Business/Technology-Home/PowerManagement.aspx)

• 

Koomey, Jonathan, Kenneth G. Brill, W. Pitt Turner, John R. Stanley, and Bruce Taylor. 2007. A simple model for determining true total cost of ownership for data centers. Santa Fe, NM: The Uptime Institute. September. (http:// www.upsite.com/cgi-bin/admin/admin.pl?admin=view_whitepapers)

• 

Koomey, Jonathan. 2008. "Worldwide electricity used in data centers." Environmental Research Letters. vol. 3, no. 034008. September 23. .

• 

Taylor, Cody, and Jonathan Koomey. 2008. Estimating energy use and greenhouse gas emissions of Internet advertising. Working paper for IMC2. February 14. . 35