《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 14 Towards Renewable Energy Powered Sustainable and Green Cloud Datacenters

TowardsRenewableEnergyPoweredSustainableand High-Performance CloudDatacentersProf.Tao LiIDEAL (lIntelligent Design of Efficient Architectures Laboratory)Electricaland ComputerEngineeringUniversityofFlorida
Towards Renewable Energy Powered Sustainable and High-Performance Cloud Datacenters Prof. Tao Li IDEAL (Intelligent Design of Efficient Architectures Laboratory) Electrical and Computer Engineering University of Florida

"Energy is the Big Issue'-Discover Magazine,2010Engincering~20KwWperrackPowerBudget~5kwperrack~1000Watts~100WattsHigh-densityExascale ComputingofficeLow-densityPersonalFacilitiesBladeServerServerRacksWorkstationComputer“Exascale computers...need a dozen nuclearpowerstationstorunit."-E&T:theEngineering&TechnologyMagazineIDEAL2RESEARCHI
“Energy is the Big Issue” - Discover Magazine, 2010 ~100 Watts ~20KW per rack ~5KW per rack ~1000 Watts Power Budget 2 ~100 Watts Personal Computer Office Workstation Low-density Server Racks High-density Blade Server Exascale Computing Facilities “Exascale computers . need a dozen nuclear power stations to run it.” - E&T: the Engineering & Technology Magazine

DataCenterPowerDemandGrowsngineerinAnnualDataCenterEnergyDemand(%ofTotal U.S.ElectricityGenerated)[2,3]10%8%5%3%0%2000200520102015Growth of data center powerdemand (Mw),2011-2012 (1)[0, 50][50,100][750,1000][500,750][250,500][100,250][1] DcD Industry Census2011:Forecasting Energy Demand,http://www.dcd-intelligence.com/IDEAL[2] Conferencereport:TheFuture oftheData Centre,http://www.information-age.com3RESEARCHI[3] Ken Brill,The Economic Meltdown of Moore's LawandtheGreenData Center
Data Center Power Demand Grows [2, 3] 5% 8% 10% Annual Data Center Energy Demand (% of Total U.S. Electricity Generated) 3 [750, 1000) [500 , 750) [250 , 500) [100 , 250) [50 , 100) [0, 50) [1] DCD Industry Census 2011: Forecasting Energy Demand, http://www.dcd-intelligence.com/ [2] Conference report: The Future of the Data Centre, http://www.information-age.com [3] Ken Brill, The Economic Meltdown of Moore’s Law and the Green Data Center • Growth of data center power demand (MW), 2011 – 2012 [1] 0% 3% 5% 2000 2005 2010 2015

TheEnergyExpenditureProblemEnginceringThe 3-Year Energy ExpenditureHistoricalElectricityPricesinUK(%ofTotalITEquipmentCost)[3] [1,2]180200%a160140150%120A100%1008050%60AverageComparedto20050%-Benchmarkfor2005402020002003200620092012Escalating powerfootprint drives data center cost upNeedalternativepowerprovisioningsolutions[1] Conference report:TheFuture ofthe Data Centre,http://www.information-age.comIDEAL[2] Ken Brill, The Economic Meltdown of Moore's Law and the Green Data Center4[3]https://www.gov.uk/government/organisations/department-of-energy-climate-changeRESEARCH
50% 100% 150% 200% The 3-Year Energy Expenditure (% of Total IT Equipment Cost) The Energy Expenditure Problem [1,2] [3] 80 100 120 140 160 180 Retail Price Index Historical Electricity Prices in UK 4 0% 50% 2000 2003 2006 2009 2012 [3] https://www.gov.uk/government/organisations/department-of-energy-climate-change • Escalating power footprint drives data center cost up – Need alternative power provisioning solutions [1] Conference report: The Future of the Data Centre, http://www.information-age.com [2] Ken Brill, The Economic Meltdown of Moore’s Law and the Green Data Center 20 40 60 80 Retail Price Index Average Compared to 2005 Benchmark for 2005

WorldwideEnvironmentalConcernsEnginceringTheGlobalMeanSeaLevelTrendinGlobalcO2Emissions(MSL)HasIncreasedby2.5[2]FromFossilFuelCombustion[1]InchesSince1992366032(ww)isw4528302415C201992199720022007201219901995200020052010A 1MW data center →> 10~15 Kt cO2 yearly [3]800MW projected capacitygrowth→2.2%increase of annual COz emission in the U.S.[1] Carbon Dioxide Information Analysis Center,http://cdiac.ornl.goy/trends/emis/meth_reg.htmlIDEAL[2] http://sealevel.colorado.edu/5[3http://www.apcmedia.com/salestools/WTOL-7DJLN9_R1_EN.swfRESEARCHI
Worldwide Environmental Concerns 15 30 45 60 MSL (mm) The Global Mean Sea Level (MSL) Has Increased by 2.5 Inches Since 1992 24 28 32 36 CO2 Emissions (Gt) Trend in Global CO2 Emissions From Fossil Fuel Combustion [1] [2] 5 0 15 1992 1997 2002 2007 2012 MSL (mm) • A 1MW data center → 10~15 Kt CO2 yearly • 800 MW projected capacity growth → 2.2% increase of annual CO2 emission in the U.S. [1] Carbon Dioxide Information Analysis Center, http://cdiac.ornl.gov/trends/emis/meth_reg.html [2] http://sealevel.colorado.edu/ [3 http://www.apcmedia.com/salestools/WTOL-7DJLN9_R1_EN.swf 20 24 1990 1995 2000 2005 2010 CO2 Emissions (Gt) [3]

AdditionalDrivingPower of GreenITnginceringGovernment regulationsand initiativesEnvironmentalProtectionAgency(http://www.epa.gov/)EU EmissionTrading Scheme(http:l/ec.europa.eu/clima/policies)Renewables Portfolio Standard (Wikipedia:renewable portfolio standard)Non-profit organizations campaignfor sustainabilityGreenpeace(http://www.greenpeace.orgRegional GreenhouseGasInitiative(http://www.rggi.org)UnionofConcerned Scientists(http://ucsusa.org)The green energy concept has mass appeal among consumers []SolarEnergy69%新新茂曲FavorableWind Energy65%VeryFavorableBio-fuel54%[1] Energy&EnvironmentConsumerSurvey,PikeResearch,2012IDEAL6RESEARCH
Additional Driving Power of Green IT • Non-profit organizations campaign for sustainability • Government regulations and initiatives Environmental Protection Agency (http://www.epa.gov/ ) EU Emission Trading Scheme (http://ec.europa.eu/clima/policies ) Renewables Portfolio Standard ( Wikipedia: renewable portfolio standard ) 6 Greenpeace (http://www.greenpeace.org) Regional Greenhouse Gas Initiative (http://www.rggi.org) Union of Concerned Scientists (http://ucsusa.org) [1] Energy & Environment Consumer Survey , Pike Research, 2012 [1] • The green energy concept has mass appeal among consumers Bio-fuel Wind Energy Solar Energy Favorable Very Favorable 69% 65% 54%

EvaluatingITSustainabilityEngineeringClimateTotal SustainabilityScoreinEnergyPoliticalSolutionsImpactAdvocacy2011(TotalScorein2010)Full Score:1o0 pt.!61130Google47 (33)9HP221445 (41)021IBM2546 (42)11515Intel31 (27)61112Microsoft29 (31)444Oracle12 (n/a)[source:greenpeace.org]" IT... must offer cutting-edge, technological solutions that canachieve significant reductionsinglobal warmingpollution..-GreenpeaceInternalIDEAL7RESEARCH!
Evaluating IT Sustainability Total Sustainability Score in 2011 (Total Score in 2010) Full Score: 100 pt. ! Climate Solutions Energy Impact Political Advocacy Google 47 (33) 11 6 30 HP 45 (41) 22 14 9 IBM 46 (42) 25 21 0 7 Intel 31 (27) 11 15 5 Microsoft 29 (31) 11 12 6 Oracle 12 (n/a) 4 4 4 “ IT. must offer cutting-edge, technological solutions that can achieve significant reductions in global warming pollution .” - Greenpeace Internal [source: greenpeace.org]

RenewableEnergyDrivenDesignngineerinsepuimerosero[Source:NREL]CharacteristicsofRenewableEnergySupplyFederalsupportConversion EfficiencyLowImprovedbrandreputationHighInitial CostAbundantenergyresourceTime-varyingEnergyGenerationLoweroperatingcostSupplyAvailabilityTime-varyingConversionEnergyGridITas anHarvestingEfficiencyIntegrationEnabler?IDEAL8RESEARCHI
Renewable Energy Driven Design • Federal support Biomass Solar Wind Ocean Geothermal Hydrogen [Source: NREL] Characteristics of Renewable Energy Supply 8 • Federal support • Improved brand reputation • Abundant energy resource • Lower operating cost Characteristics of Renewable Energy Supply Conversion Efficiency Low Initial Cost High Energy Generation Time-varying Supply Availability Time-varying Conversion Efficiency IT as an Enabler? Grid Integration Energy Harvesting

TowardsRenewableEnergyPoweredDataCentersngineerliHPExpandsRenewableEnergyUseinitsGlobalOperations福hp2000PALOALTO.Calif..Noy.27.2007We've invested overAbetterwebGoogle$915million inBetterfortheenvironmentrenewable energyYAHOO!Key Scientific Challenges Blog Series:Green ComputinghpPostedFebruary1st,2010at 10:02am byReganClarkIBMRolls OutFirst SolarArrayDesignedForHigh-VoltageDataTBMCenters and Industrial UseBangalorepilotcould lead toall-new solutions forpowering servers andGougretechgearinemergingeconomiesMicrosoftlookingtotestgrid-independentdata centerMicrosoftRadicalintegrationfromenergysourcethroughmicroprocessorproposedETOZebaMicrosoftAPPLEREACHES1OO%RENEWABLEENERGYACROSSALIDATA CENTERSGreenpeace approves, but pushes for more pressure on legislatorsPMachaeapyYevgeniySverdlik-DatacenterDynamicsOtherWorld ComputingIDEAL9RESEARCHI
Towards Renewable Energy Powered Data Centers 200 7 9 201 3

BigGreenWSC+DataEnergyDiversified PowerVery LargeScaleProvisioningComputingAutonomicGreenComputing
Big Data WSC Green Energy Very Large Scale Computing Diversified Power Provisioning 10 Autonomic Green Computing
按次数下载不扣除下载券;
注册用户24小时内重复下载只扣除一次;
顺序:VIP每日次数-->可用次数-->下载券;
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 15 GPGPU Architecture and Programming Paradigm.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 12 Shared Memory Multiprocessor.pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 1 Instruction Set Architecture(Introduction).pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 0 Introduction and Performance Evaluation.pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 3 Pipelining.pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 2 Instruction Set Architecture(Microarchitecture Implementation).pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 7 Multiprocessors.pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 4 Spectualtive Execution.pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 6 Memory Hierarchy and Cache.pdf
- 《现代计算机体系结构》课程教学课件(留学生版)Lecture 5 Out of Order Execution.pdf
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第4章 基于统计决策的概率分类法.ppt
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第1章 绪论、第2章 聚类分析.ppt
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第3章 判别函数及几何分类法.ppt
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第7章 模糊模式识别法.ppt
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第6章 句法模式识别.ppt
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第5章 特征选择与特征提取.ppt
- 武汉理工大学:《模式识别》课程教学资源(PPT课件)第8章 神经网络模式识别法.ppt
- 武汉理工大学:《模式识别》课程教学资源(实验指导,共五个实验).pdf
- 武汉理工大学:《模式识别》课程授课教案(讲义)第8章 神经网络在模式识别中的应用.pdf
- 武汉理工大学:《模式识别》课程授课教案(讲义)第7章 模糊模式识别.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 11 Multi-core and Multi-threading.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 10 Out of Order and Speculative Execution.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 13 An Introduction to Cloud Data Centers.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 09 Case Study- Jave Branch Prediction Optimization.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 07 Instruction Decode.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 08 Instruction Fetch and Branch Predictioin.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 06 Scoreboarding and Tomasulo.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 04 Memory Data Prefetching.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 05 Core Pipelining.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 02 Memory Hierarchy and Caches.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 03 Main Memory and DRAM.pdf
- 《现代计算机体系结构》课程教学课件(英文讲稿)Lecture 01 Introduction and Performance Evaluation-new.pdf
- 东北大学:某学院计算机科学与技术专业《智能信息系统开发》课程教学大纲.pdf
- 东北大学:某学院计算机科学与技术专业《软件工程综合实践》课程教学大纲.pdf
- 东北大学:某学院计算机科学与技术专业《创新创业设计基础》课程教学大纲.pdf
- 东北大学:某学院计算机科学与技术专业《工程领导力》课程教学大纲.pdf
- 东北大学:某学院计算机科学与技术专业《高等数学建模》课程教学大纲(二).pdf
- 东北大学:某学院计算机科学与技术专业《数据库原理》课程教学大纲.pdf
- 东北大学:某学院计算机科学与技术专业《物理建模》课程教学大纲 A.pdf
- 东北大学:某学院计算机科学与技术专业《Java程序设计基础》课程教学大纲.pdf