{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

iccd08-keynote - Towards More Sustainable Computer Design...

Info icon This preview shows pages 1–16. Sign up to view the full content.

View Full Document Right Arrow Icon
owards More Sustainable Towards More Sustainable Computer Design Fred Chong Director, Greenscale Center for Energy-Efficient Computing Director, Computer Engineering UC Santa Barbara
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Skyrocketing Energy Consumption [adapted from Pacala & Socolow Science 2004]
Image of page 2
atacenter xample Datacenter Example EPA report, Public Law 109-231, 2007
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
nergy Savings Energy Savings 15 years: • In 15 years: – 300 fewer power stations liminate 7 8 billion etric tons of arbon – eliminate 7.8 billion metric tons of carbon – save 10,000 TWh of electricity
Image of page 4
hree Challenges Three Challenges atacenters: design for energy- Datacenters: design for energy proportional computing • Embedded Systems: design for silicon use reuse • Cultural change: multidisciplinary green thinking
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
art I: Efficient Datacenters Part I:
Image of page 6
rver = SUV Server SUV = ore precisely: More precisely: – 80 billion terawatt-hr / yr = 6 million SUVs in arbon production (10 mpg 11K miles/yr) carbon production (10 mpg, 11K miles/yr)
Image of page 7

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
arehouse Computing Warehouse Computing = • Google at Columbia River Gorge => ydroelectric power hydroelectric power • $30B annual energy bill worldwide nergy starting to cost more than capital • Energy starting to cost more than capital expenditures
Image of page 8
First Steps: Airflow • Airflow modeling omputation scheduling Computation scheduling – Move floor tiles ot/cold aisles baffles – Hot/cold aisles, baffles • Next steps – Active louvers – Real-time CFD Image courtesy Flomerics – Precise control Goal: air cooling with the efficiency of liquid cooling (1000X efficiency)
Image of page 9

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Problem: Lightly Utilized Servers [Barosso and Hoetzle, Computer 2007]
Image of page 10
nergy- fficiency at Light Load Energy Efficiency at Light Load [Barosso and Hoetzle, Computer 2007]
Image of page 11

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
nergy- roportional Computing Energy Proportional Computing [Barosso and Hoetzle, Computer 2007]
Image of page 12
wo Strategies Two Strategies erver consolidation Server consolidation – Move load to increase utilization – But need to turn off idle servers and routers • Energy-proportional hardware ot just CPU Not just CPU – Disruptive technologies in omputation / cooling computation / cooling
Image of page 13

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
rver Consolidation Server Consolidation uch work in virtualization and Much work in virtualization and consolidation ut shutting down machines is • But shutting down machines is problematic achines are slow to come back up – Machines are slow to come back up – Machines don’t come back in consistent tate state – Fault-tolerant software re-replicates data and ervices services
Image of page 14
nergy- roportional Clusters Energy Proportional Clusters • Fast state save/restore hardware – Non-volatile storage and high bandwidth to processor and router tate state
Image of page 15

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 16
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern