25
Energy Efficient Data Centers Update on LBNL data center energy efficiency projects June 23, 2005 Bill Tschudi Lawrence Berkeley National Laboratory [email protected]

Energy Efficient Data Centers Update on LBNL data center energy efficiency projects June 23, 2005 Bill Tschudi Lawrence Berkeley National Laboratory [email protected]

Embed Size (px)

Citation preview

  • Energy Efficient Data CentersJune 23, 2005

    Bill TschudiLawrence Berkeley National Laboratory

    [email protected]

    Update on LBNL data center energy efficiency projects

  • LBNLs energy research related to data centersEnergy research roadmapCase studies and energy benchmarkingBest practice identificationSelf benchmarking protocolInvestigate efficiency of power supplies in IT equipment Investigate efficiency of UPS systemsMetrics for computing performance vs. energy Technology transferDemonstration projects

  • Data center efficiency opportunityMany efficiency ideas have been identified through industry feedbackCase studies are helping to identify best practices

  • Data center efficiency resourcesASHRAE Thermal Guidelines for Data Processing EnvironmentsASHRAE Power Trends and Cooling ApplicationsIn preparation: ASHRAE Design Considerations for Data Center and Communications Equipment Centers which includes a chapter on energy efficiency

  • Case studies/benchmarksCalifornia Storage device and router Mfgs. BanksWeb hosting facilitiesInternet service providerState tax centerFederal facilities New York Recovery center (hosting)Financial institution

  • IT equipment load intensities Data collected in 1999 through 2003 showed that electrical power intensity for IT equipment alone was on the order of 25 Watts/sf.

    Current data suggests that load intensities are rising through compaction and/or due to rising equipment power consumption.

  • 2003 IT equipment loads from LBNL case studiesAverage 27 +/-(W/sf of electrically active floor space using Uptime definition)

  • 2003 projections if fully loaded

    Projected Average 44(W/sf of electrically active floor space using Uptime definition)

  • Distribution of computer load intensities reported to Uptime Institute

    Chart1

    000Number of facilities353848

    0.05042179770.01019428710.008063137601.551.721.86

    0.05042179770.01019428710.0080631376022.922.425.3

    0.09564045380.12990081380.0451981865035.49538.52847.058

    0.09564045380.12990081380.0451981865

    0.21975564820.15904066930.0666998868

    0.21975564820.15904066930.0666998868

    0.24447525780.16749938660.0887391295

    0.24447525780.16749938660.0887391295

    0.27033258990.18978554810.1991503603

    0.27033258990.18978554810.1991503603

    0.29358479590.22339079520.2260274856

    0.29358479590.22339079520.2260274856

    0.3129777950.26206288030.2774756789

    0.3129777950.26206288030.2774756789

    0.38020685870.26965089870.3099432463

    0.38020685870.26965089870.3099432463

    0.4017007660.34873646670.3282293672

    0.4017007660.34873646670.3282293672

    0.47207795990.36622038010.3745569306

    0.47207795990.36622038010.3745569306

    0.50042406020.39177603340.3955904313

    0.50042406020.39177603340.3955904313

    0.51807621450.45536677460.455795192

    0.51807621450.45536677460.455795192

    0.60450434730.47497789740.5186521874

    0.60450434730.47497789740.5186521874

    0.62508936940.54013461450.5353160051

    0.62508936940.54013461450.5353160051

    0.63438637320.5616625570.54966839

    0.63438637320.5616625570.54966839

    0.64916707070.56863222760.5793004207

    0.64916707070.56863222760.5793004207

    0.66694592590.62108396770.586758823

    0.66694592590.62108396770.586758823

    0.67608843210.68366705270.5941795973

    0.67608843210.68366705270.5941795973

    0.68908238790.71317640160.6050954729

    0.68908238790.71317640160.6050954729

    0.71790490970.72126853950.6197338305

    0.71790490970.72126853950.6197338305

    0.73076893240.74243806180.6469952987

    0.73076893240.74243806180.6469952987

    0.77601926370.75415286650.6534237695

    0.77601926370.75415286650.6534237695

    0.78248036460.7657505290.675080282

    0.78248036460.7657505290.675080282

    0.80393354670.77902198490.6857773778

    0.80393354670.77902198490.6857773778

    0.82112867250.79085684580.7008737216

    0.82112867250.79085684580.7008737216

    0.84713985580.81003087080.7116788635

    0.84713985580.81003087080.7116788635

    0.86169494810.82800841330.7152599717

    0.86169494810.82800841330.7152599717

    0.86924852130.87463218220.7276234493

    0.86924852130.87463218220.7276234493

    0.89914347590.89397346990.7375357331

    0.89914347590.89397346990.7375357331

    0.90552377260.90709572970.7589137986

    0.90552377260.90709572970.7589137986

    0.94241313550.92296121540.76228204

    0.94241313550.92296121540.76228204

    0.95974401240.9261164790.7632845567

    0.95974401240.9261164790.7632845567

    0.97731277680.95271767040.767853668

    0.97731277680.95271767040.767853668

    0.98041307090.96404433220.7880652663

    0.98041307090.96404433220.7880652663

    10.97278512330.8146558814

    0.97278512330.8146558814

    0.9948754650.8277375158

    0.9948754650.8277375158

    0.99861993640.8621402362

    0.99861993640.8621402362

    10.9037718283

    0.9037718283

    0.9216112514

    0.9216112514

    0.9342795157

    0.9342795157

    0.9363065884

    0.9363065884

    0.9490533339

    0.9490533339

    0.9755251522

    0.9755251522

    0.9881063345

    0.9881063345

    0.9928711113

    0.9928711113

    0.9955529108

    0.9955529108

    0.9987270993

    0.9987270993

    1

    1999

    2000

    2001

    Source: Uptime Institute, 2002.

    1999

    2000

    2001

    Computer room UPS power (Watts/square foot)

    Fraction of total floor area in sample

    Data and results

    1999200020011999SORTED BY W/SFCumulativeData for plot2000SORTED BY W/SFCumulativeData for plot20012001SORTED BY W/SFCumulativeData for plot2001SORTED BY business type

    SortOrderBusTypeW/sfsfSortOrderBusTypeW/sfsfSortOrderBusTypeW/sfsfSortOrderBusTypeW/sfsfW/sf% of total sf% of total sf7.70.0%SortOrderBusTypeW/sfsfW/sf% of total sf% of total sf7.00.0%SortOrderBusTypeW/sfsfW/sf% of total sf% of total sf9.20.0%SortOrderBusTypeW/sfsfW/sf% of total sf

    1stock mkt supp30.4331871stock mkt supp35.3331871stock mkt supp41.03318714oil7.7780007.75.0%5.0%7.75.0%57oil7.0174927.01.0%1.0%7.01.0%13financial9.2150009.20.8%0.8%9.20.8%24aircraft mfg*36.83760036.82.0%

    2aircraft mfg*46.6570662aircraft mfg*51.2379042aircraft mfg*45.54924640telecom9.6699519.64.5%9.6%9.65.0%18govt data proc*10.920540010.912.0%13.0%10.91.0%14oil9.4690839.43.7%4.5%9.40.8%2aircraft mfg*45.54924645.52.6%

    3telecom21.21337003telecom15.11357003telecom18.99571018govt data proc*11.319200011.312.4%22.0%9.69.6%51financial11.05000011.02.9%15.9%10.913.0%43airline9.8400009.82.2%6.7%9.44.5%43airline9.8400009.82.2%

    7financial17.21040004consumer prod72.164254consumer prod77.1590570airline reserv*11.83824011.82.5%24.4%11.39.6%13financial11.11451411.10.8%16.7%11.013.0%51financial12.84100012.82.2%8.9%9.84.5%5airline33.5626633.50.3%

    9oil34.6402385airline41.654145airline33.5626651financial14.44000014.42.6%27.0%11.322.0%70airline reserv*12.03824012.02.2%19.0%11.015.9%18govt data proc*13.120540013.111.0%19.9%9.86.7%70airline reserv*21.22670021.21.4%

    11stock mkt supp22.1143827financial20.41118006online storage & access39.32433615insurance*15.03597015.02.3%29.4%11.822.0%14oil12.75766212.73.4%22.3%11.115.9%48financial14.45000014.42.7%22.6%12.86.7%42airline reserv*25.22723225.21.5%

    12financial17.43325011stock mkt supp22.9119597financial20.911200043airline16.33000016.31.9%31.3%11.824.4%40telecom14.16635614.13.9%26.2%11.116.7%3telecom18.99571018.95.1%27.7%12.88.9%54computer vendor*66.1886466.10.5%

    14oil7.77800012financial19.3336508financial35.418657financial17.210400017.26.7%38.0%14.424.4%25consumer prod14.91302014.90.8%27.0%12.016.7%36govt data proc*19.06040019.03.2%31.0%13.18.9%56consumer prod29.82010129.81.1%

    15insurance*15.03597013financial11.1145149oil25.84028812financial17.43325017.42.1%40.2%14.427.0%3telecom15.113570015.17.9%34.9%12.019.0%12financial19.53401819.51.8%32.8%13.119.9%74consumer prod36.5850036.50.5%

    16telecom19.910887014oil12.75766211stock mkt supp25.81195916telecom19.910887019.97.0%47.2%15.027.0%43airline15.33000015.31.7%36.6%12.719.0%16telecom20.78618420.74.6%37.5%14.419.9%38consumer prod42.42356742.41.3%

    18govt data proc*11.319200016telecom17.110911312financial19.53401842airline reserv*20.34385020.32.8%50.0%15.029.4%42airline reserv*15.94385015.92.6%39.2%12.722.3%39telecom20.93912920.92.1%39.6%14.422.6%4consumer prod77.1590577.10.3%

    19pkg delivery*54.32717817financial32.33084713financial9.21500041financial21.12730721.11.8%51.8%16.329.4%16telecom17.110911317.16.4%45.5%14.122.3%7financial20.911200020.96.0%45.6%18.922.6%23equip mfg*21.03100021.01.7%

    21telecom49.62681018govt data proc*10.920540014oil9.4690833telecom21.213370021.28.6%60.5%16.331.3%12financial19.33365019.32.0%47.5%14.126.2%50telecom20.911693420.96.3%51.9%18.927.7%13financial9.2150009.20.8%

    24aircraft mfg*32.92660019pkg delivery*40.12722316telecom20.78618439telecom21.23184421.22.1%62.5%17.231.3%7financial20.411180020.46.5%54.0%14.926.2%23equip mfg*21.03100021.01.7%53.5%19.027.7%51financial12.84100012.82.2%

    25consumer prod30.0999520financial28.52277218govt data proc*13.120540011stock mkt supp22.11438222.10.9%63.4%17.238.0%39telecom21.23693921.22.2%56.2%14.927.0%70airline reserv*21.22670021.21.4%55.0%19.031.0%48financial14.45000014.42.7%

    27oil38.01168524aircraft mfg*31.63290019pkg delivery*59.02340533insurance*22.42286522.41.5%64.9%17.438.0%11stock mkt supp22.91195922.90.7%56.9%15.127.0%40telecom22.25512522.23.0%57.9%19.531.0%12financial19.53401819.51.8%

    31financial39.9987025consumer prod14.91302022financial43.0377138consumer prod22.72750322.71.8%66.7%17.440.2%53telecom23.39000023.35.2%62.1%15.134.9%46stock mkt supp23.41387523.40.7%58.7%19.532.8%7financial20.911200020.96.0%

    32telecom23.64458730airline50.91499823equip mfg*21.03100046stock mkt supp22.81414322.80.9%67.6%19.940.2%50telecom23.510738423.56.3%68.4%15.334.9%29mgmt consultg23.71380523.70.7%59.4%20.732.8%41financial23.82030723.81.1%

    33insurance*22.42286532telecom24.05063424aircraft mfg*36.83760056consumer prod23.42010123.41.3%68.9%19.947.2%32telecom24.05063424.03.0%71.3%15.336.6%41financial23.82030723.81.1%60.5%20.737.5%47financial27.91990027.91.1%

    38consumer prod22.72750338consumer prod49.51943527oil29.72808432telecom23.64458723.62.9%71.8%20.347.2%46stock mkt supp26.61388526.60.8%72.1%15.936.6%42airline reserv*25.22723225.21.5%62.0%20.937.5%35financial32.51844032.51.0%

    39telecom21.23184439telecom21.23693928stock mkt supp30.3666247financial23.91990023.91.3%73.1%20.350.0%48financial27.03632427.02.1%74.2%15.939.2%32telecom25.65071525.62.7%64.7%20.939.6%8financial35.4186535.40.1%

    40telecom9.66995140telecom14.16635629mgmt consultg23.71380552financial23.97000023.94.5%77.6%21.150.0%56consumer prod27.12010127.11.2%75.4%17.139.2%11stock mkt supp25.81195925.80.6%65.3%20.939.6%52financial40.27744840.24.2%

    41financial21.12730741financial30.62030732telecom25.65071525consumer prod30.0999530.00.6%78.2%21.151.8%47financial27.81990027.81.2%76.6%17.145.5%9oil25.84028825.82.2%67.5%20.945.6%22financial43.0377143.00.2%

    42airline reserv*20.34385042airline reserv*15.94385034pkg delivery*32.8397701stock mkt supp30.43318730.42.1%80.4%21.251.8%20financial28.52277228.51.3%77.9%19.345.5%47financial27.91990027.91.1%68.6%20.945.6%18govt data proc*13.120540013.111.0%

    43airline16.33000043airline15.33000035financial32.51844024aircraft mfg*32.92660032.91.7%82.1%21.260.5%41financial30.62030730.61.2%79.1%19.347.5%27oil29.72808429.71.5%70.1%20.951.9%36govt data proc*19.06040019.03.2%

    44stock mkt supp35.62251644stock mkt supp37.82251636govt data proc*19.0604009oil34.64023834.62.6%84.7%21.260.5%24aircraft mfg*31.63290031.61.9%81.0%20.447.5%56consumer prod29.82010129.81.1%71.2%21.051.9%29mgmt consultg23.71380523.70.7%

    46stock mkt supp22.81414346stock mkt supp26.61388537mgmt consultg66.1498944stock mkt supp35.62251635.61.5%86.2%21.262.5%17financial32.33084732.31.8%82.8%20.454.0%28stock mkt supp30.3666230.30.4%71.5%21.053.5%37mgmt consultg66.1498966.10.3%

    47financial23.91990047financial27.81990038consumer prod42.42356727oil38.01168538.00.8%86.9%22.162.5%52financial34.38000034.34.7%87.5%21.254.0%45oil32.02300032.01.2%72.8%21.253.5%14oil9.4690839.43.7%

    49online storage & access58.0479648financial27.03632439telecom20.93912955telecom38.44624638.43.0%89.9%22.163.4%1stock mkt supp35.33318735.31.9%89.4%21.256.2%35financial32.51844032.51.0%73.8%21.255.0%9oil25.84028825.82.2%

    51financial14.44000049online storage & access72.6236840telecom22.25512531financial39.9987039.90.6%90.6%22.463.4%44stock mkt supp37.82251637.81.3%90.7%22.956.2%34pkg delivery*32.83977032.82.1%75.9%22.255.0%27oil29.72808429.71.5%

    52financial23.97000050telecom23.510738441financial23.8203072aircraft mfg*46.65706646.63.7%94.2%22.464.9%19pkg delivery*40.12722340.11.6%92.3%22.956.9%5airline33.5626633.50.3%76.2%22.257.9%45oil32.02300032.01.2%

    55telecom38.44624651financial11.05000042airline reserv*25.22723221telecom49.62681049.61.7%96.0%22.764.9%5airline41.6541441.60.3%92.6%23.356.9%8financial35.4186535.40.1%76.3%23.457.9%6online storage & access39.32433639.31.3%

    56consumer prod23.42010152financial34.38000043airline9.84000019pkg delivery*54.32717854.31.8%97.7%22.766.7%55telecom42.54564442.52.7%95.3%23.362.1%74consumer prod36.5850036.50.5%76.8%23.458.7%49online storage & access80.7236880.70.1%

    58financial72.83030053telecom23.39000044stock mkt supp43.42371349online storage & access58.0479658.00.3%98.0%22.866.7%38consumer prod49.51943549.51.1%96.4%23.562.1%24aircraft mfg*36.83760036.82.0%78.8%23.758.7%34pkg delivery*32.83977032.82.1%

    70airline reserv*11.83824055telecom42.54564445oil32.02300058financial72.83030072.82.0%100.0%22.867.6%30airline50.91499850.90.9%97.3%23.568.4%55telecom38.24946738.22.7%81.5%23.759.4%19pkg delivery*59.02340559.01.3%

    56consumer prod27.12010146stock mkt supp23.41387523.467.6%2aircraft mfg*51.23790451.22.2%99.5%24.068.4%6online storage & access39.32433639.31.3%82.8%23.859.4%46stock mkt supp23.41387523.40.7%

    57oil7.01749247financial27.91990023.468.9%4consumer prod72.1642572.10.4%99.9%24.071.3%53telecom39.66400039.63.4%86.2%23.860.5%11stock mkt supp25.81195925.80.6%

    70airline reserv*12.03824048financial14.45000023.668.9%49online storage & access72.6236872.60.1%100.0%26.671.3%52financial40.27744840.24.2%90.4%25.260.5%28stock mkt supp30.3666230.30.4%

    49online storage & access80.7236823.671.8%26.672.1%1stock mkt supp41.03318741.01.8%92.2%25.262.0%1stock mkt supp41.03318741.01.8%

    50telecom20.911693423.971.8%27.072.1%38consumer prod42.42356742.41.3%93.4%25.662.0%44stock mkt supp43.42371343.41.3%

    51financial12.84100023.973.1%27.074.2%22financial43.0377143.00.2%93.6%25.664.7%3telecom18.99571018.95.1%

    52financial40.27744823.973.1%27.174.2%44stock mkt supp43.42371343.41.3%94.9%25.864.7%16telecom20.78618420.74.6%

    53telecom39.66400023.977.6%27.175.4%2aircraft mfg*45.54924645.52.6%97.6%25.865.3%39telecom20.93912920.92.1%

    54computer vendor*66.1886430.077.6%27.875.4%19pkg delivery*59.02340559.01.3%98.8%25.865.3%50telecom20.911693420.96.3%

    55telecom38.24946730.078.2%27.876.6%54computer vendor*66.1886466.10.5%99.3%25.867.5%40telecom22.25512522.23.0%

    56consumer prod29.82010130.478.2%28.576.6%37mgmt consultg66.1498966.10.3%99.6%27.967.5%32telecom25.65071525.62.7%

    70airline reserv*21.22670030.480.4%28.577.9%4consumer prod77.1590577.10.3%99.9%27.968.6%55telecom38.24946738.22.7%

    74consumer prod36.5850032.980.4%30.677.9%49online storage & access80.7236880.70.1%100.0%29.768.6%53telecom39.66400039.63.4%

    32.982.1%30.679.1%29.770.1%

    34.682.1%31.679.1%29.870.1%

    22.9154695022.4171586325.3186031822.91546950100.0%34.684.7%22.41715863100.0%31.681.0%25.31860318100.0%29.871.2%25.31860318100.0%

    35.684.7%32.381.0%30.371.2%

    Data point #10 was removed because the source apparently was not reporting data for the same facility in the three years shown here.35.686.2%32.382.8%30.371.5%

    Create a subset of the data centers for which we have data for all three years38.086.2%34.382.8%32.071.5%

    38.086.9%34.387.5%32.072.8%

    19992000200138.486.9%35.387.5%32.572.8%

    1stock mkt supp30.433187.01stock mkt supp35.333187.01stock mkt supp41.033187.038.489.9%35.389.4%32.573.8%

    2aircraft mfg*46.657066.02aircraft mfg*51.237904.02aircraft mfg*45.549246.039.989.9%37.889.4%32.873.8%

    3telecom21.2133700.03telecom15.1135700.03telecom18.995710.039.990.6%37.890.7%32.875.9%

    7financial17.2104000.07financial20.4111800.07financial20.9112000.046.690.6%40.190.7%33.575.9%

    11stock mkt supp22.114382.011stock mkt supp22.911959.011stock mkt supp25.811959.046.694.2%40.192.3%33.576.2%

    12financial17.433250.012financial19.333650.012financial19.534018.049.694.2%41.692.3%35.476.2%

    14oil7.778000.014oil12.757662.014oil9.469083.049.696.0%41.692.6%35.476.3%

    16telecom19.9108870.016telecom17.1109113.016telecom20.786184.054.396.0%42.592.6%36.576.3%

    18govt data proc*11.3192000.018govt data proc*10.9205400.018govt data proc*13.1205400.054.397.7%42.595.3%36.576.8%

    19pkg delivery*54.327178.019pkg delivery*40.127223.019pkg delivery*59.023405.058.097.7%49.595.3%36.876.8%

    24aircraft mfg*32.926600.024aircraft mfg*31.632900.024aircraft mfg*36.837600.058.098.0%49.596.4%36.878.8%

    32telecom23.644587.032telecom24.050634.032telecom25.650715.072.898.0%50.996.4%38.278.8%

    38consumer prod22.727503.038consumer prod49.519435.038consumer prod42.423567.072.8100.0%50.997.3%38.281.5%

    39telecom21.231844.039telecom21.236939.039telecom20.939129.051.297.3%39.381.5%

    40telecom9.669951.040telecom14.166356.040telecom22.255125.051.299.5%39.382.8%

    41financial21.127307.041financial30.620307.041financial23.820307.072.199.5%39.682.8%

    42airline reserv*20.343850.042airline reserv*15.943850.042airline reserv*25.227232.072.199.9%39.686.2%

    43airline16.330000.043airline15.330000.043airline9.840000.072.699.9%40.286.2%

    44stock mkt supp35.622516.044stock mkt supp37.822516.044stock mkt supp43.423713.072.6100.0%40.290.4%

    46stock mkt supp22.814143.046stock mkt supp26.613885.046stock mkt supp23.413875.041.090.4%

    47financial23.919900.047financial27.819900.047financial27.919900.041.092.2%

    49online storage & access58.04796.049online storage & access72.62368.049online storage & access80.72368.042.492.2%

    51financial14.440000.051financial11.050000.051financial12.841000.042.493.4%

    52financial23.970000.052financial34.380000.052financial40.277448.043.093.4%

    55telecom38.446246.055telecom42.545644.055telecom38.249467.043.093.6%

    56consumer prod23.420101.056consumer prod27.120101.056consumer prod29.820101.043.493.6%

    70airline reserv*11.838240.070airline reserv*12.038240.070airline reserv*21.226700.043.494.9%

    45.594.9%

    20.8135921721.5135667324.3128843945.597.6%

    59.097.6%

    W/sfW/sfW/sfPASTED AND SORTEDW/sfW/sfW/sf59.098.8%

    19992000200119992000200166.198.8%

    1Stock mkt supp30354124Aircraft mfg*33323766.199.3%

    2Aircraft mfg*4751462Aircraft mfg*47514666.199.3%

    3Telecom21151943Airline16151066.199.6%

    7Financial17202170Airline reserv*12122177.199.6%

    11Stock mkt supp22232642Airline reserv*20162577.199.9%

    12Financial17192038Consumer prod23494280.799.9%

    14Oil813956Consumer prod23273080.7100.0%

    16Telecom20172151Financial141113

    18Govt data proc*1111137Financial172021

    19Pkg delivery*54405912Financial171920

    24Aircraft mfg*33323741Financial213124

    32Telecom24242647Financial242828

    38Consumer prod23494252Financial243440

    39Telecom21212118Govt data proc*111113

    40Telecom10142214Oil8139

    41Financial21312449Online storage & access587381

    42Airline reserv*20162519Pkg delivery*544059

    43Airline16151011Stock mkt supp222326

    44Stock mkt supp36384346Stock mkt supp232723

    46Stock mkt supp2327231Stock mkt supp303541

    47Financial24282844Stock mkt supp363843

    49Online storage & access58738140Telecom101422

    51Financial14111316Telecom201721

    52Financial2434403Telecom211519

    55Telecom38433839Telecom212121

    56Consumer prod23273032Telecom242426

    70Airline reserv*12122155Telecom384338

    The following data point was removed because the institution reporting the data apparently was doing so for different facilities (the year 2000 data is for a different facility than is the year 1999 data)

    10financial22.1143820022.9119590025.811959

    Table for PPT

    Number of facilitiesTotal floor areaComputer room power densityTotal power use

    Million square feetW/square footMW

    1999351.5522.935

    2000381.7222.439

    2001481.8625.347

    Graph

    80.0070.0090.00

    80.0570.0190.01

    100.05110.0190.01

    100.10110.1390.050.751.51.141.2825

    110.10110.13100.051314000000

    110.22110.16100.07

    120.22110.16130.07

    120.24110.17130.09

    140.24120.17130.09

    140.27120.19130.20

    150.27130.19140.20

    150.29130.22140.23

    160.29140.22190.23

    160.31140.26190.28

    170.31150.26190.28

    170.38150.27190.31

    170.38150.27200.31

    170.40150.35200.33

    200.40150.35210.33

    200.47150.37210.37

    200.47160.37210.37

    200.50160.39210.40

    210.50170.39210.40

    210.52170.46210.46

    210.52190.46210.46

    210.60190.47210.52

    210.60200.47210.52

    210.63200.54210.54

    220.63210.54210.54

    220.63210.56210.55

    220.63230.56220.55

    220.65230.57220.58

    230.65230.57230.58

    230.67230.62230.59

    230.67230.62240.59

    230.68230.68240.59

    230.68240.68240.59

    230.69240.71240.61

    240.69270.71250.61

    240.72270.72250.62

    240.72270.72260.62

    240.73270.74260.65

    240.73270.74260.65

    240.78270.75260.65

    300.78280.75260.65

    300.78280.77260.68

    300.78290.77280.68

    300.80290.78280.69

    330.80310.78300.69

    330.82310.79300.70

    350.82320.79300.70

    350.85320.81300.71

    360.85320.81300.71

    360.86320.83300.72

    380.86340.83320.72

    380.87340.87320.73

    380.87350.87330.73

    380.90350.89330.74

    400.90380.89330.74

    400.91380.91330.76

    470.91400.91340.76

    470.94400.92340.76

    500.94420.92350.76

    500.96420.93350.76

    540.96430.93360.76

    540.98430.95360.77

    580.98490.95370.77

    580.98490.96370.79

    730.98510.96380.79

    731.00510.97380.81

    510.97390.81

    510.99390.83

    720.99400.83

    721.00400.86

    731.00400.86

    731.00400.90

    410.90

    410.92

    420.92

    420.93

    430.93

    430.94

    430.94

    430.95

    460.95

    460.98

    590.98

    590.99

    660.99

    660.99

    660.99

    661.00

    771.00

    771.00

    811.00

    811.00

    Graph

    000Number of facilities353848

    00001.551.721.86

    000022.922.425.3

    000035.49538.52847.058

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    00

    00

    00

    00

    00

    00

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    1999

    2000

    2001

    Source: Uptime Institute, 2002.

    1999

    2000

    2001

    Computer room UPS power (Watts/square foot)

    Fraction of total floor area in sample

    000Number of facilities353848

    00001.551.721.86

    000022.922.425.3

    000035.49538.52847.058

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    000

    00

    00

    00

    00

    00

    00

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    1999

    2000

    2001

    1999

    2000

    2001

    Computer room UPS power (Watts/square foot)

    Fraction of total floor area in sample

  • 2005 IT equipment benchmarksLBNL NERSC supercomputerAverage 52 w/sf

  • Electrical power conversion is a big opportunityEvery conversion of AC voltage, AC to DC, DC to AC, or DC voltage conversion results in loss of electrical power and corresponding heat that must be removed from the data center. Minimizing this conversion loss has a magnifying effect that allows all facility systems to use less energy and consequently the infrastructure systems can be downsized. Saving say 10% on the conversion loss could result in 20% or more saving for the facility.

  • How many times do data centers convert AC and DC?AC voltage conversions

  • Measured UPS efficiency

    Chart1

    78.5949389.867.586.78185.584.689.647.3

    4.1.1

    6.1.1

    6.2.1

    6.2.2

    8.2.1

    8.2.2

    8.2.3

    8.2.4

    8.2.5

    9.1.1

    9.1.2

    Load Factor (%)

    Efficiency (%)

    UPS Efficiency

    Sheet1

    Sheet1

    78.5949389.867.586.78185.584.689.647.3

    4.1.1

    6.1.1

    6.2.1

    6.2.2

    8.2.1

    8.2.2

    8.2.3

    8.2.4

    8.2.5

    9.1.1

    9.1.2

    Load Factor (%)

    Efficiency (%)

    UPS Efficiency

    Sheet2

    Sheet3

  • Measured UPS losses

  • Measured UPS losses

  • Electricity use in a serverBased on a typical dual processor 450W 2U Server; Approximately 160W out of 450W (35%) are losses in the power conversion process (Source: Brian Griffith: INTEL)

    Chart2

    130.5

    31.5

    31.5

    72

    40.5

    85.5

    27

    31.5

    Sheet1

    PS Design GuideNumber of Output BusOutput DC VoltagesWattage SizeApplicationRedundancyEfficiency RecommendationForm Factor

    EPS Rev 1.043.3 V, 5V, 12V, -12V300W to 400WEntry Level Server Computernon-Redundant1no recommendationPS2

    EPS 12V Rev 2.173.3 V, 5V, 12V1, 12V2, 12V3, 12V4, -12V and 5Vsb450W-650WEntry Level Server Computernon-Redundant1450W-68%; 550W-70%, 65W-72%PS2

    EPS 1U, Rev 2.153.3V, 5V, 12V, -12V and 5 VSB125W to 500Wfor 1U server systemnon-Redundant1125W - 65% to 500W -75%1U Rack Chasis

    EPS 2U, Rev 2.153.3V, 5V, 12V, -12V and 5 VSB400W to 700Wfor 2U server systemnon-Redundant1480W - 68% to 650W -72%2U Rack Chasis

    ERP12V Rev 2.13 to 53.3V (optional) 5V (optinal) 12V, -12V and 5 VSB45oW to 750WFor Pedestal serversRedundant Power450W-65%, 650W-70%Pedestal System

    ERP12V Rev 2.03 to 53.3V (optional) 5V (optinal) 12V, -12V and 5 VSB350W - 700WFor 2U rack mount serversRedundant Power350W-70% to 650W 82%for 2U Rack System

    A D2D Rev 1.013.3V, 5V, 12V, 2.5V (VID Family of Voltages: 1.3-3.5)Approx 100WFor DC-DC ConvertersRedundancy required if operating in N+15V, 12V, >3.3V >=85%; VID

  • Power supply opportunityBased on one case study approximately 4335 KW of a total of 8500 kW was IT load. Assuming a 65% existing baseline efficiency, the savings opportunity using 90% efficient conversion process is approximately 1300kW not including any savings from HVAC

    Chart3

    2380

    85

    170

    340

    4335

    1190

    Sheet1

    PS Design GuideNumber of Output BusOutput DC VoltagesWattage SizeApplicationRedundancyEfficiency RecommendationForm Factor

    EPS Rev 1.043.3 V, 5V, 12V, -12V300W to 400WEntry Level Server Computerno recommendationPS2

    EPS 12V Rev 2.173.3 V, 5V, 12V1, 12V2, 12V3, 12V4, -12V and 5Vsb450W-650WEntry Level Server Computer450W-68%; 550W-70%, 65W-72%PS2

    EPS 1U, Rev 2.153.3V, 5V, 12V, -12V and 5 VSB125W to 500Wfor 1U server system125W - 65% to 500W -75%1U Rack Chasis

    EPS 2U, Rev 2.153.3V, 5V, 12V, -12V and 5 VSB400W to 700Wfor 2U server system480W - 68% to 650W -72%2U Rack Chasis

    ERP12V Rev 2.13 to 53.3V (optional) 5V (optinal) 12V, -12V and 5 VSB45oW to 750WFor Pedestal serversRedundant Power450W-65%, 650W-70%Pedestal System

    ERP12V Rev 2.03 to 53.3V (optional) 5V (optinal) 12V, -12V and 5 VSB350W - 700WFor 2U rack mount serversRedundant Power350W-70% to 650W 82%for 2U Rack System

    A D2D Rev 1.013.3V, 5V, 12V, 2.5V (VID Family of Voltages: 1.3-3.5)Approx 100WFor DC-DC ConvertersRedundancy required if operating in N+15V, 12V, >3.3V >=85%; VID 3.3V >=85%; VID

  • Power supply efficiency recommendationsThe Server Systems Infrastructure group (SSI) publishes recommended minimum efficiencies for server power supplies. The LBNL project team is working with this group to see what can be done to raise the bar.

  • Power supply efficiency today

  • Measured power supply efficiency

  • Energy efficiency opportunitySpecifiers of UPSs or IT equipment can have a huge impact on energy use by requiring higher efficiencies. Testing data shows that higher efficiencies can be obtained you have to ask for it. Facility and IT professionals by working together can optimize overall power conversions. Additional costs (if any) for more efficient conversions will have a very short payback or may be entirely justified by reductions in infrastructure.

  • Effectiveness of HVAC systems

    Chart3

    0.29

    0.31

    0.25

    0.5

    0.36

    0.31

    0.21

    0.35

    0.32

    0.54

    0.224

    0.46

    Data Center Identifier

    % of total load

    HVAC (as a % of total load)

    Summary Data

    Data Center Load Characterization Project

    All Sites Summary

    FacilityComputer Load Density Based on Gross Floor AreaHVAC Power Consumption as Percent of Total Data Center Power Consumption

    W/sf

    12429%

    23431%

    34425%

    4.14.150%

    4.54.836%

    535.331%

    6.164.521%

    6.24835%

    71332%

    8.18.554%

    8.214.522%

    912.446%

    1029

    Summary Data

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    Data Center Identifier

    W/Sq.ft.

    Computer Load Density

    CHART_Computer Load Density

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    0

    Data Center Identifier

    % of total load

    HVAC

    CHART_HVAC Power

    24

    34

    44

    4.1

    4.8

    35.3

    64.5

    48

    13

    8.5

    14.5

    12.4

    All values are based on the respective data center gross floor area.

    W/sf

    Facility Number

    Watts per Square Foot

    Computer Load Density

    0.29

    0.31

    0.25

    0.5

    0.36

    0.31

    0.21

    0.35

    0.32

    0.54

    0.224

    0.46

    All values are shown as a percent of the respective data center total power consumption.

    Facility Number

    Data Center Load Characterization ProjectHVAC Power Consumption

  • Index of performanceThe Uptime Institute proposed a metric to evaluate the total efficiency of infrastructure systems: Index of performance = building systems KW UPS output(i.e. ratio of building systems to IT equipment load)

  • Data Center AData Center BLook at the end-use

  • LBNL high-tech buildings website:http://hightech.lbl.gov