Ultra Efficient HPC Data
          Center
          Natural Low Energy Cooling Conceptual Design




10/8/10                                                  1
Ultra-Efficient HPC Data Centre - Gary Bernstein, McGill University
Project Funding

    Canada's Advanced Research and Innovation
     Network (CANARIE)
    Canada California Strategic Innovation Partnership
     (CCSIP)
     •    ISTP Canada
     •    University of California
     •    McGill University
Site Selection

    Three candidate locations in Quebec
     •    McDonald Campus of McGill University in St. Anne de Bellevue
     •    Campus of the Institut de recherche d’Hydro-Quebec (IREQ) in
          Varennes
     •    IREQ campus in Shawinigan

    McDonald Campus of McGill University selected as
     site for project
    All enjoy
              Cold climate
              Renewable energy resource (hydroelectric)
              Inexpensive electricity
The System Approach: An Overview

    Goals: Most Efficient Class One Data Center
    Climate Evaluation
    Define Loads and How to Best Serve Them
     •    Water cooled equipment
     •    Medium temperature chilled water (65F, 75F)

    Optimize Heat Rejection for Climate and Loads
     •    Evaporative free cooling – Primary cooling
     •    Seasonal ice storage – Top up cooling

    Backup Approach
    Results
Goals


    Provide ASHRAE TC9.9 Class 1 Datacenter
    No compressor based cooling
    Lower construction cost
    Lower operating cost
    Best efficiency
    Environmentally friendly
          Construction materials
          Recycle heat, water
Proposed Annual Electrical Costs
Comparison


  $10,000,000

   $9,000,000

   $8,000,000

   $7,000,000
                                          $5 Million/yr Annual
   $6,000,000                               Savings Target
   $5,000,000

   $4,000,000

   $3,000,000

   $2,000,000

   $1,000,000

          $0
                San Diego (1.35 PUE)   Montreal (1.06 PUE)
                   at $0.09/kWh          at $0.05/kWh
Aerial Perspective – “Farm” at
McDonald Campus
         Cooling Towers and
     Mechanical Infrastructure                     20,000 SF Phase 1
                                                   8 MW IT Load
 VA
 Hospital

                                                           Fuel Tanks


Cooling Ice Pond


Stormwater Detension


            Office, Shipping/
                   Receiving          Electrical
                                 Infrastructure
Climate: Free Cooling Analysis with 65F CHWS


                                            0.030
                                                       Data Source: Government of Canada - National Climate Data & Information Archive
                                            0.028
                                                       Data Set: WMO #71627, Montreal/Pierre Elliott Trudeau Airport, Typical Year
                                            0.026
                                                       Elevation: 118 feet
 Humidity Ratio (lbs H2O per lbs dry air)




                                            0.024
                                                       Air Pressure: 14.633224 psia
                                            0.022

                                            0.020                                                                                        Auxillary Cooling
                                                                                                                                             80 hrs/yr
                                            0.018

                                            0.016                                                           Partial Free Cooling
                                                                                                                1234 hrs/yr
                                            0.014

                                            0.012

                                            0.010

                                            0.008
                                                                                           Full Free Cooling
                                            0.006                                             7446 hrs/yr
                                            0.004

                                            0.002

                                            0.000
                                                    -30 -25 -20 -15 -10 -5    0    5   10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

                                                                                             Dry Bulb Temperature (F)
Climate: Free Cooling Analysis With 75F CHWS


                                            0.030
                                                       Data Source: Government of Canada - National Climate Data & Information Archive
                                            0.028
                                                       Data Set: Montreal/Pierre Elliott Trudeau Intl Airport, Typical Year
                                            0.026      Elevation: 118 feet                                                                   Auxillary Cooling
                                                       Air Pressure: 14.633224 psia                                                               0 hrs/yr
 Humidity Ratio (lbs H2O per lbs dry air)




                                            0.024

                                            0.022

                                            0.020                                                                              Partial Free Cooling

                                            0.018                                                                                   114 hrs/yr

                                            0.016

                                            0.014

                                            0.012

                                            0.010

                                            0.008
                                                                                              Full Free Cooling
                                            0.006                                                8646 hrs/yr
                                            0.004

                                            0.002

                                            0.000
                                                    -30 -25 -20 -15 -10 -5      0    5   10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

                                                                                                Dry Bulb Temperature (F)
Load Strategies
    Climate analysis shows higher temperature chilled
     water offers many more hours of free cooling
    Highly concentrated heat loads
     •    A single high density rack
          can put off as much
          waste heat as a
          VW Beetle (40kW)
     •    Air exiting racks
          typically exceeds
          90F
Load Heat Collection Strategies

    Higher temperature Chilled Water Supply (CHWS)
     offers many more hours of free cooling: Design to
     use 75F and 65F CHWS
     •    Direct water based cooling most efficient
     •    Hot aisle / cold aisle for minority of load
Primary Cooling Strategies: Medium
Temp. Cooling Water and Free Cooling
Primary Cooling Strategies: Medium
 Temp. Cooling Water and Free Cooling
    Design to cool with 65F/18C and 75F/24C water
     •    90% of 65F load served with cooling tower provided free cooling;
          99.3% of 75F load
     •    590,000 ton-hrs (2,100 MWh) Top up Cooling Required
Supplemental Cooling: Seasonal Ice
     Storage Slush Pond System
     Fill in winter with plowed snow collection
     Melt water cools data center
Ultra-Efficient HPC Data Centre - Gary Bernstein, McGill University
Slush Pond

     Paved collection basin, 75,000 ft3 (2,100 m3)
            Drive-in slope on one side for plow loading
            Lightweight, waterproof insulating cover or roof to protect
             from warm rains
            Extensive drain system to collect meltwater
            Berms for sides, or dig into ground
Sundsvall, Sweden, Snow Storage - Empty
Sundsvall, Sweden, Snow Storage - Full
Snow dump overruns in Montreal, 2008-09
Slush Pond System – Pumping and
     Filtration
     Mature waste water handling technology
            Remove gravel, wood, grit from melt water
            Remove oils and road chemicals prior to release as required
            Filter

     Select heat exchangers for highly corrosive fluid
            Maintain complete separation between pond water and
             building loop water

     Integrate settling tank to also serve as emergency
      storage
Key approaches

    Keep storage pond simple
           Leverage local snow removal program if possible
           Collect snow dumpage fees?

    Provide appropriate maintenance
           Provide for pile grooming, drain clearing, filter cleaning, etc

    Use in lieu of chillers to save cost
           Consider emergency chiller rental for backup

    Design properly
           Simple concept but careful design required
Office Approaches
    Much lower load
    Design for comfort and optimal use of medium
     temperature water
Backup – Do Not Invest in Chillers 'Just
in Case'!

Pay for it only
when (if) you
ever need it


Design for
portable
air-cooled
chillers
to connect in
an emergency
Results

    McGill-USCD HPC Data Center PUE Itemization
                                                     Fans; 1.5%


                                                    CRAH Fans; 0.0%

                                                      Humidifier; 0.0%

                                                     CHW Plant; 2.1%

                                                      Transformer Loss;
                                                            0.5%
                                                      UPS Loss; 0.6%

       Racks; 94.0%                                   PDU Loss; 1.0%



                                                Data Center
                                                Lights; 0.2%


  Power Usage Effectiveness (PUE) = Total Energy / Rack Energy = 1.06
Results

Supply Temperatures                              Annual Energy Use    Mechanical Cooling Needed Water Usage
                      Hours of Free
                      Cooling / year     PUE                                  Additional Load at
  Air      Water                                            Cost                                 Evaporation +
 Cooled    Cooled
                                                Energy
                                                       ( $0.058/kWh)  Hours Extreme Weather Carry Over
                                                                     per Year (wetbulb = 68.7°F)
°C   °F   °C    °F    hrs/yr   % of yr          MWh/yr        $                       tons          gallons

23.9 75.0 23.9 75.0 8,532       97%      1.06    74,567     $4,325,000 228             0          30,100,000

More Related Content

PDF
IBM Global Financing
PDF
IBM zEnterprise: Healthcare
PDF
Business Optimization Thru Information Analytics
PPT
Dc energy efficiency presentation for psu lecture - ashok bhatla - final
PDF
IBM zEnterprise: Banking
PDF
Smarter Planet: Government - Public Safety
PDF
Cyber Summit 2016: Technology, Education, and Democracy
PPTX
Cyber Summit 2016: Understanding Users' (In)Secure Behaviour
IBM Global Financing
IBM zEnterprise: Healthcare
Business Optimization Thru Information Analytics
Dc energy efficiency presentation for psu lecture - ashok bhatla - final
IBM zEnterprise: Banking
Smarter Planet: Government - Public Safety
Cyber Summit 2016: Technology, Education, and Democracy
Cyber Summit 2016: Understanding Users' (In)Secure Behaviour

More from Cybera Inc. (20)

PPT
Cyber Summit 2016: Insider Threat Indicators: Human Behaviour
PPTX
Cyber Summit 2016: Research Data and the Canadian Innovation Challenge
PPTX
Cyber Summit 2016: Knowing More and Understanding Less in the Age of Big Data
PDF
Cyber Summit 2016: Privacy Issues in Big Data Sharing and Reuse
PPTX
Cyber Summit 2016: Establishing an Ethics Framework for Predictive Analytics ...
PDF
Cyber Summit 2016: The Data Tsunami vs The Network: How More Data Changes Eve...
PDF
Cyber Summit 2016: Issues and Challenges Facing Municipalities In Securing Data
PPT
Cyber Summit 2016: Using Law Responsibly: What Happens When Law Meets Technol...
PPT
Privacy, Security & Access to Data
PDF
Do Universities Dream of Big Data
PDF
Predicting the Future With Microsoft Bing
PPTX
Analytics 101: How to not fail at analytics
PPTX
Are MOOC's past their peak?
PPTX
Opening the doors of the laboratory
PPTX
Open City - Edmonton
PPTX
Unlocking the power of healthcare data
PPT
Checking in on Healthcare Data Analytics
PPTX
Open access and open data: international trends and strategic context
PPTX
I didn't know i was a geomatics company
PPTX
Integrating Geospatial into the Everyday
Cyber Summit 2016: Insider Threat Indicators: Human Behaviour
Cyber Summit 2016: Research Data and the Canadian Innovation Challenge
Cyber Summit 2016: Knowing More and Understanding Less in the Age of Big Data
Cyber Summit 2016: Privacy Issues in Big Data Sharing and Reuse
Cyber Summit 2016: Establishing an Ethics Framework for Predictive Analytics ...
Cyber Summit 2016: The Data Tsunami vs The Network: How More Data Changes Eve...
Cyber Summit 2016: Issues and Challenges Facing Municipalities In Securing Data
Cyber Summit 2016: Using Law Responsibly: What Happens When Law Meets Technol...
Privacy, Security & Access to Data
Do Universities Dream of Big Data
Predicting the Future With Microsoft Bing
Analytics 101: How to not fail at analytics
Are MOOC's past their peak?
Opening the doors of the laboratory
Open City - Edmonton
Unlocking the power of healthcare data
Checking in on Healthcare Data Analytics
Open access and open data: international trends and strategic context
I didn't know i was a geomatics company
Integrating Geospatial into the Everyday
Ad

Recently uploaded (20)

PDF
Five Habits of High-Impact Board Members
PDF
Enhancing emotion recognition model for a student engagement use case through...
PPTX
O2C Customer Invoices to Receipt V15A.pptx
PPTX
The various Industrial Revolutions .pptx
PPTX
Benefits of Physical activity for teenagers.pptx
PDF
Architecture types and enterprise applications.pdf
PDF
Zenith AI: Advanced Artificial Intelligence
PDF
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
PPTX
Tartificialntelligence_presentation.pptx
PDF
Hybrid model detection and classification of lung cancer
PPTX
observCloud-Native Containerability and monitoring.pptx
DOCX
search engine optimization ppt fir known well about this
PDF
Microsoft Solutions Partner Drive Digital Transformation with D365.pdf
PPTX
MicrosoftCybserSecurityReferenceArchitecture-April-2025.pptx
PDF
1 - Historical Antecedents, Social Consideration.pdf
PDF
Univ-Connecticut-ChatGPT-Presentaion.pdf
PPT
What is a Computer? Input Devices /output devices
PDF
Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor
PDF
A comparative study of natural language inference in Swahili using monolingua...
PDF
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
Five Habits of High-Impact Board Members
Enhancing emotion recognition model for a student engagement use case through...
O2C Customer Invoices to Receipt V15A.pptx
The various Industrial Revolutions .pptx
Benefits of Physical activity for teenagers.pptx
Architecture types and enterprise applications.pdf
Zenith AI: Advanced Artificial Intelligence
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
Tartificialntelligence_presentation.pptx
Hybrid model detection and classification of lung cancer
observCloud-Native Containerability and monitoring.pptx
search engine optimization ppt fir known well about this
Microsoft Solutions Partner Drive Digital Transformation with D365.pdf
MicrosoftCybserSecurityReferenceArchitecture-April-2025.pptx
1 - Historical Antecedents, Social Consideration.pdf
Univ-Connecticut-ChatGPT-Presentaion.pdf
What is a Computer? Input Devices /output devices
Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor
A comparative study of natural language inference in Swahili using monolingua...
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
Ad

Ultra-Efficient HPC Data Centre - Gary Bernstein, McGill University

  • 1. Ultra Efficient HPC Data Center Natural Low Energy Cooling Conceptual Design 10/8/10 1
  • 3. Project Funding   Canada's Advanced Research and Innovation Network (CANARIE)   Canada California Strategic Innovation Partnership (CCSIP) •  ISTP Canada •  University of California •  McGill University
  • 4. Site Selection   Three candidate locations in Quebec •  McDonald Campus of McGill University in St. Anne de Bellevue •  Campus of the Institut de recherche d’Hydro-Quebec (IREQ) in Varennes •  IREQ campus in Shawinigan   McDonald Campus of McGill University selected as site for project   All enjoy   Cold climate   Renewable energy resource (hydroelectric)   Inexpensive electricity
  • 5. The System Approach: An Overview   Goals: Most Efficient Class One Data Center   Climate Evaluation   Define Loads and How to Best Serve Them •  Water cooled equipment •  Medium temperature chilled water (65F, 75F)   Optimize Heat Rejection for Climate and Loads •  Evaporative free cooling – Primary cooling •  Seasonal ice storage – Top up cooling   Backup Approach   Results
  • 6. Goals   Provide ASHRAE TC9.9 Class 1 Datacenter   No compressor based cooling   Lower construction cost   Lower operating cost   Best efficiency   Environmentally friendly   Construction materials   Recycle heat, water
  • 7. Proposed Annual Electrical Costs Comparison $10,000,000 $9,000,000 $8,000,000 $7,000,000 $5 Million/yr Annual $6,000,000 Savings Target $5,000,000 $4,000,000 $3,000,000 $2,000,000 $1,000,000 $0 San Diego (1.35 PUE) Montreal (1.06 PUE) at $0.09/kWh at $0.05/kWh
  • 8. Aerial Perspective – “Farm” at McDonald Campus Cooling Towers and Mechanical Infrastructure 20,000 SF Phase 1 8 MW IT Load VA Hospital Fuel Tanks Cooling Ice Pond Stormwater Detension Office, Shipping/ Receiving Electrical Infrastructure
  • 9. Climate: Free Cooling Analysis with 65F CHWS 0.030 Data Source: Government of Canada - National Climate Data & Information Archive 0.028 Data Set: WMO #71627, Montreal/Pierre Elliott Trudeau Airport, Typical Year 0.026 Elevation: 118 feet Humidity Ratio (lbs H2O per lbs dry air) 0.024 Air Pressure: 14.633224 psia 0.022 0.020 Auxillary Cooling 80 hrs/yr 0.018 0.016 Partial Free Cooling 1234 hrs/yr 0.014 0.012 0.010 0.008 Full Free Cooling 0.006 7446 hrs/yr 0.004 0.002 0.000 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Dry Bulb Temperature (F)
  • 10. Climate: Free Cooling Analysis With 75F CHWS 0.030 Data Source: Government of Canada - National Climate Data & Information Archive 0.028 Data Set: Montreal/Pierre Elliott Trudeau Intl Airport, Typical Year 0.026 Elevation: 118 feet Auxillary Cooling Air Pressure: 14.633224 psia 0 hrs/yr Humidity Ratio (lbs H2O per lbs dry air) 0.024 0.022 0.020 Partial Free Cooling 0.018 114 hrs/yr 0.016 0.014 0.012 0.010 0.008 Full Free Cooling 0.006 8646 hrs/yr 0.004 0.002 0.000 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Dry Bulb Temperature (F)
  • 11. Load Strategies   Climate analysis shows higher temperature chilled water offers many more hours of free cooling   Highly concentrated heat loads •  A single high density rack can put off as much waste heat as a VW Beetle (40kW) •  Air exiting racks typically exceeds 90F
  • 12. Load Heat Collection Strategies   Higher temperature Chilled Water Supply (CHWS) offers many more hours of free cooling: Design to use 75F and 65F CHWS •  Direct water based cooling most efficient •  Hot aisle / cold aisle for minority of load
  • 13. Primary Cooling Strategies: Medium Temp. Cooling Water and Free Cooling
  • 14. Primary Cooling Strategies: Medium Temp. Cooling Water and Free Cooling   Design to cool with 65F/18C and 75F/24C water •  90% of 65F load served with cooling tower provided free cooling; 99.3% of 75F load •  590,000 ton-hrs (2,100 MWh) Top up Cooling Required
  • 15. Supplemental Cooling: Seasonal Ice Storage Slush Pond System   Fill in winter with plowed snow collection   Melt water cools data center
  • 17. Slush Pond   Paved collection basin, 75,000 ft3 (2,100 m3)   Drive-in slope on one side for plow loading   Lightweight, waterproof insulating cover or roof to protect from warm rains   Extensive drain system to collect meltwater   Berms for sides, or dig into ground
  • 18. Sundsvall, Sweden, Snow Storage - Empty
  • 19. Sundsvall, Sweden, Snow Storage - Full
  • 20. Snow dump overruns in Montreal, 2008-09
  • 21. Slush Pond System – Pumping and Filtration   Mature waste water handling technology   Remove gravel, wood, grit from melt water   Remove oils and road chemicals prior to release as required   Filter   Select heat exchangers for highly corrosive fluid   Maintain complete separation between pond water and building loop water   Integrate settling tank to also serve as emergency storage
  • 22. Key approaches   Keep storage pond simple   Leverage local snow removal program if possible   Collect snow dumpage fees?   Provide appropriate maintenance   Provide for pile grooming, drain clearing, filter cleaning, etc   Use in lieu of chillers to save cost   Consider emergency chiller rental for backup   Design properly   Simple concept but careful design required
  • 23. Office Approaches   Much lower load   Design for comfort and optimal use of medium temperature water
  • 24. Backup – Do Not Invest in Chillers 'Just in Case'! Pay for it only when (if) you ever need it Design for portable air-cooled chillers to connect in an emergency
  • 25. Results McGill-USCD HPC Data Center PUE Itemization Fans; 1.5% CRAH Fans; 0.0% Humidifier; 0.0% CHW Plant; 2.1% Transformer Loss; 0.5% UPS Loss; 0.6% Racks; 94.0% PDU Loss; 1.0% Data Center Lights; 0.2% Power Usage Effectiveness (PUE) = Total Energy / Rack Energy = 1.06
  • 26. Results Supply Temperatures Annual Energy Use Mechanical Cooling Needed Water Usage Hours of Free Cooling / year PUE Additional Load at Air Water Cost Evaporation + Cooled Cooled Energy ( $0.058/kWh) Hours Extreme Weather Carry Over per Year (wetbulb = 68.7°F) °C °F °C °F hrs/yr % of yr MWh/yr $ tons gallons 23.9 75.0 23.9 75.0 8,532 97% 1.06 74,567 $4,325,000 228 0 30,100,000