A Computer Scientist Looks at the Energy Problem Randy H. Katz - - PowerPoint PPT Presentation

▶
a computer scientist looks at the energy problem
SMART_READER_LITE
LIVE PREVIEW

A Computer Scientist Looks at the Energy Problem Randy H. Katz - - PowerPoint PPT Presentation

Nov 25, 2023 162 likes •688 views

A Computer Scientist Looks at the Energy Problem Randy H. Katz University of California, Berkeley Usenix Technical Symposium San Diego, CA June 19, 2009 Energy permits things to exist; information, to behave purposefully. W. Ware, 1997

slide-1
SLIDE 1

A Computer Scientist Looks at the Energy Problem

Randy H. Katz

University of California, Berkeley Usenix Technical Symposium San Diego, CA June 19, 2009

“Energy permits things to exist; information, to behave purposefully.”

  • W. Ware, 1997
slide-2
SLIDE 2

Agenda

  • The Big Picture
  • IT as an Energy Consumer
  • IT as an Efficiency Enabler
  • Summary and Conclusions

2

slide-3
SLIDE 3

Energy “Spaghetti” Chart

10-8-2008 3

Quads (1015 BTUs)

slide-4
SLIDE 4

Electricity is the Heart of the Energy Economy

4

slide-5
SLIDE 5

The Big Switch: Clouds + Smart Grids

5

Computing as a Utility Computing in the Utility Large-scale industrialization

  • f computing

Energy Efficient Computing Embedded Intelligence in Civilian Infrastructures

slide-6
SLIDE 6

Energy + Information Flow = Third Industrial Revolution

“The coming together of distributed communication technologies and distributed renewable energies via an open access, intelligent power grid, represents “power to the people”. For a younger generation that’s growing up in a less hierarchical and more networked world, the ability to produce and share their

  • wn energy, like they produce and

share their own information, in an

  • pen access intergrid, will seem

both natural and commonplace.”

6

Jeremy Rifkin

slide-7
SLIDE 7

Agenda

  • The Big Picture
  • IT as an Energy Consumer
  • IT as an Efficiency Enabler
  • Summary and Conclusions

7

slide-8
SLIDE 8

2020 IT Carbon Footprint

8

820m tons CO2 360m tons CO2 260m tons CO2 2007 Worldwide IT carbon footprint: 2% = 830 m tons CO2 Comparable to the global aviation industry Expected to grow to 4% by 2020

slide-9
SLIDE 9

2020 IT Carbon Footprint

9

“SMART 2020: Enabling the Low Carbon Economy in the Information Age”, The Climate Group

USA China Telecoms DC PCs Datacenters: Owned by single entity interested in reducing opex billion tons CO2

slide-10
SLIDE 10

2020 IT Carbon Footprint

10

Projected Savings

“SMART 2020: Enabling the Low Carbon Economy in the Information Age”, The Climate Group

slide-11
SLIDE 11

Energy Proportional Computing

11 Figure 1. Average CPU utilization of more than 5,000 servers during a six-month period. Servers are rarely completely idle and seldom operate near their maximum utilization, instead operating most of the time at between 10 and 50 percent of their maximum

It is surprisingly hard to achieve high levels

  • f utilization of typical

servers (and your home PC or laptop is even worse) “The Case for Energy-Proportional Computing,” Luiz André Barroso, Urs Hölzle, IEEE Computer December 2007

slide-12
SLIDE 12

Energy Proportional Computing

12 Figure 2. Server power usage and energy efficiency at varying utilization levels, from idle to peak performance. Even an energy-efficient server still consumes about half its full power when doing virtually no work.

“The Case for Energy-Proportional Computing,” Luiz André Barroso, Urs Hölzle, IEEE Computer December 2007 Doing nothing well … NOT!

Energy Efficiency = Utilization/Power

slide-13
SLIDE 13

Energy Proportional Computing

13 Figure 4. Power usage and energy efficiency in a more energy-proportional server. This server has a power efficiency of more than 80 percent of its peak value for utilizations of 30 percent and above, with efficiency remaining above 50 percent for utilization levels as low as 10 percent.

“The Case for Energy-Proportional Computing,” Luiz André Barroso, Urs Hölzle, IEEE Computer December 2007 Design for wide dynamic power range and active low power modes Doing nothing VERY well

Energy Efficiency = Utilization/Power

slide-14
SLIDE 14

14

Internet Datacenters

slide-15
SLIDE 15

Energy Use In Datacenters

LBNL Michael Patterson, Intel

Datacenter Energy Overheads

slide-16
SLIDE 16

DC Infrastructure Energy Efficiencies

16

Cooling (Air + Water movement) + Power Distribution

slide-17
SLIDE 17

Containerized Datacenter Mechanical-Electrical Design

17

Google’s Containerized Datacenter Microsoft Chicago Datacenter

slide-18
SLIDE 18

Power Usage Effectiveness Rapidly Approaching 1!

18

Bottom-line: the frontier of DC energy efficiency IS the IT equipment Doing nothing well becomes incredibly important

slide-19
SLIDE 19

19

Datacenter Power

Peak Power %

slide-20
SLIDE 20

20

Datacenter Power

Transformer Main Supply

ATS Switch Board UPS UPS STS PDU STS PDU Panel Panel Generator …

1000 kW 200 kW 50 kW

Rack

Circuit 2.5 kW

  • X. Fan, W-D Weber, L. Barroso, “Power Provisioning for a

Warehouse-sized Computer,” ISCA’07, San Diego, (June 2007).

  • Typical structure 1MW

Tier-2 datacenter

  • Reliable Power

– Mains + Generator – Dual UPS

  • Units of Aggregation

– Rack (10-80 nodes) – PDU (20-60 racks) – Facility/Datacenter

slide-21
SLIDE 21

21

Nameplate vs. Actual Peak

  • X. Fan, W-D Weber, L. Barroso, “Power Provisioning for a

Warehouse-sized Computer,” ISCA’07, San Diego, (June 2007).

Component CPU Memory Disk PCI Slots Mother Board Fan System Total Peak Power 40 W 9 W 12 W 25 W 25 W 10 W Count 2 4 1 2 1 1 Total 80 W 36 W 12 W 50 W 25 W 10 W 213 W Nameplate peak 145 W Measured Peak (Power-intensive workload) In Google’s world, for given DC power budget, deploy as many machines as possible

slide-22
SLIDE 22

22

Typical Datacenter Power

Power-aware allocation of resources can achieve higher levels of utilization – harder to drive a cluster to high levels

  • f utilization than an individual rack
  • X. Fan, W-D Weber, L. Barroso, “Power Provisioning for a

Warehouse-sized Computer,” ISCA’07, San Diego, (June 2007).

Racks can be driven to high utilization/95% power Clusters driven to modest utilization/67% power

slide-23
SLIDE 23

23

Better to have one computer at 50% utilization than five computers at 10% utilization: Save $ via Consolidation (& Save Power)

“Power” of Consolidation: Keep Fewer Machines More Busy

  • SPECpower:

– Two 3.0-GHz Xeons, 16 GB DRAM, 1 Disk – One 2.4-GHz Xeon, 8 GB DRAM, 1 Disk

  • 50% utilization 

85% Peak Power

  • 10%  65% Peak Power
  • Save 75% power if

consolidate & turn off

1 computer @ 50% = 225 W

  • v. 5 computers @ 10% = 870 W
slide-24
SLIDE 24

24

Atoms are Quite Better at Doing Nothing Well

Measured Power in Soda Hall Machine Rooms

slide-25
SLIDE 25

Microsoft’s Chicago Modular Datacenter

25

slide-26
SLIDE 26

The Million Server Datacenter

  • 24000 sq. m housing 400 containers

– Each container contains 2500 servers – Integrated computing, networking, power, cooling systems

  • 300 MW supplied from two power

substations situated on opposite sides of the datacenter

  • Dual water-based cooling systems

circulate cold water to containers, eliminating need for air conditioned rooms

26

slide-27
SLIDE 27

Agenda

  • The Big Picture
  • IT as an Energy Consumer
  • IT as an Efficiency Enabler
  • Summary and Conclusions

27

slide-28
SLIDE 28

28

Generation Transmission Distribution

Machine Age Energy Infrastructure

slide-29
SLIDE 29

29

Accommodate 21st Century Renewable Energy Sources

slide-30
SLIDE 30

Challenge of Integrating Intermittent Sources

30

Sun and wind aren’t where the people – and the current grid – are located!

www.technologyreview.com

slide-31
SLIDE 31

California as a Testbed

31

If we do this, we will need to build a new grid to manage and move renewable energy around Day Night

www.technologyreview.com

slide-32
SLIDE 32

32

What if the Energy Infrastructure were Designed like the Internet?

  • Energy: the limited resource of the 21st Century
  • Information Age approach to Machine Age

infrastructure: bits follow current flow

– Break synchronization between sources and loads: energy storage/buffering is key – Lower cost, more incremental deployment, suitable for developing economies – Enhanced reliability and resilience to wide-area

  • utages, such as after natural disasters
  • Exploit information to match sources to loads,

manage buffers, integrate renewables, signal demand response, and take advantage of locality

slide-33
SLIDE 33

Information Overlay to the Energy Grid

33

Conventional Electric Grid Generation Transmission Distribution Load Intelligent Energy Network

Load IPS Source IPS energy subnet Intelligent Power Switch

Conventional Internet

slide-34
SLIDE 34

34

Intelligent Power Switch (IPS) Energy Network

PowerComm Interface

Energy Storage Power Generation Host Load

energy flows information flows

Intelligent Power Switch

  • PowerComm Interface: Network + Power connector
  • Scale Down, Scale Out
slide-35
SLIDE 35

35

“Doing Nothing Well”

  • Existing systems sized for peak and designed for

continuous activity

– Reclaim the idle waste – Exploit huge gap in peak-to-average power consumption

  • Continuous demand response

– Challenge “always on” assumption – Realize potential of energy-proportionality

  • From IT Equipment …

– Better fine-grained idling, faster power shutdown/ restoration – Pervasive support in operating systems and applications

  • … to the OS for the Building
slide-36
SLIDE 36

Multi-Scale Energy Internet

36

comm power

now Load profile

w $

now Price profile

w

now Actual load

w

Datacenter

Bldg Energy Network

IPS IPS IPS IPS

Internet Grid

IPS IPS Power proportional kernel Power proportional service manager

Quality- Adaptive Service

M/R Energy Net IPS IPS IPS

AHU

Chill

CT

IPS

slide-37
SLIDE 37

Smart Buildings

37

Servers / Clusters HVAC / CRU / PDU support Lighting HVAC & Plug Loads

slide-38
SLIDE 38

Physical Systems vs. Logical Use

38

slide-39
SLIDE 39

Energy Consumption Breakdown

39

slide-40
SLIDE 40

40

Cooperative Continuous Energy Reduction

Automated Control Facility Mgmt User Demand Supervisory Control Community Feedback High- fidelity visibility

slide-41
SLIDE 41

Agenda

  • The Big Picture
  • IT as an Energy Consumer
  • IT as an Efficiency Enabler
  • Summary and Conclusions

41

slide-42
SLIDE 42

Summary and Conclusions

  • Energy Consumption in IT Equipment

– Energy Proportional Computing and “Doing Nothing Well” – Management of Processor, Memory, I/O, Network to maximize performance subject to power constraints – Internet Datacenters and Containerized Datacenters: New packaging opportunities for better optimization of computing + communicating + power + mechanical

42

slide-43
SLIDE 43

43

Summary and Conclusions

  • LoCal: a scalable energy network

– Inherent inefficiencies at all levels of electrical energy distribution – Integrated energy generation and storage – IPS and PowerComm Interface – Energy matching at small, medium, large scale

  • Demand response: doing nothing well
  • Smart buildings beyond datacenters
slide-44
SLIDE 44

44

Thank You!

“We’re at the beginning of the information utility. The past is big monolithic buildings. The future looks more like a substation—the data center represents the information substation of tomorrow.” Mike Manos, Microsoft GM Datacenter Services

slide-45
SLIDE 45
slide-46
SLIDE 46
slide-47
SLIDE 47

47

slide-48
SLIDE 48

“The Big Switch” and Cloud Computing

48

“A hundred years ago, companies

stopped generating their own power with steam engines and dynamos and plugged into the newly built electric

  • grid. The cheap power pumped out by

electric utilities didn’t just change how businesses operate. It set off a chain reaction of economic and social transformations that brought the modern world into existence. Today, a similar revolution is under way. Hooked up to the Internet’s global computing grid, massive information-processing plants have begun pumping data and software code into our homes and

  • businesses. This time, it’s computing

that’s turning into a utility.”

slide-49
SLIDE 49

ACme – HiFi Metering

49

slide-50
SLIDE 50

Re-aggregation

50

slide-51
SLIDE 51

By Individual

51

slide-52
SLIDE 52

52

Energy Awareness and Adaptation

  • Export existing facilities instrumentation

into real-time feed and archival physical information base

  • Augment with extensive usage-focused

sensing

  • Create highly visible consumer feedback

and remediation guidance

  • Develop whole-building dynamic models
  • Basis for forecasting and load

sculpting