Workshop on Green and Environmental Computing - Helsinki

1,251 views
1,182 views

Published on

Published in: Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,251
On SlideShare
0
From Embeds
0
Number of Embeds
337
Actions
Shares
0
Downloads
44
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Workshop on Green and Environmental Computing - Helsinki

  1. 1. Workshop on Green and Environmental Computingp g 27th – 28th October 2010 CSC – IT Center for Science Espoo, Finland Middl hMiddleware research for green Data Centers Prof. Jordi Torres Barcelona Supercomputing Center (BSC) UPC Barcelona TechUPC – Barcelona Tech 1Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  2. 2. Contents Motivation for reduction of energy consumption What can we do for DC power efficiency? Power aware Resource Management Research at a Glance Conclusions 2Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  3. 3. The motivations are well-known by all of youby all of you Impact of computing on the environment: ICT industry produces roughly 2 to 3 percent of global 3Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing ICT industry produces roughly 2 to 3 percent of global GHG emissions.
  4. 4. And we can’t forget …g Opex Data Center 5 years (%) Data courtesy of AST IT Infrastructures Calculation for avg density 7kw/rack Infrastructure amortization counted as Opex HW and SW excluded 4Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing (www.AST-global.com) HW and SW excluded Data Center Infrastructure valued as Opex
  5. 5. Goal: Reduce DC energy usegy Has become a first-order objective in the design and operation of theoperation of the Data Centers Source: EPA report to Congress 2007 How can we reduce it? 5Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  6. 6. Contents Motivation for reduction of energy consumption What can we do for DC power efficiency? Power aware Resource Management Research at a Glance Conclusions 6Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  7. 7. What can we do for DC power efficiency?efficiency? Can we Minimize losses and thermal/coolingCan we Minimize losses and thermal/cooling overheads in a Data Center? Example: Reducing cooling costsExample: Reducing cooling costs “Best Practices to save energy at Marenostrum” Source: “El reto operacional de dirigir el tercer supercomputadorsupercomputador más grande de Europa” Sergi Girona, Director Operaciones BSC-CNS, 2009. 7Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing Foto: BSC
  8. 8. Best practice at Marenostrum Problem measured:Problem measured: to much under-floor pressure Test: Move some floor tiles Benefits observed: – Improvement of AC equipmentequipment performance – Improved the rack- 8Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing p bottom temperature Foto: BSC
  9. 9. Best practice at Marenostrum Substituting Floor Tiles – Composite • 20% opening – Metallic tile • 40% opening40% opening B fit b dBenefits observed: – Less working pressure for the Cooling components – All bladecenters reduced by 2º C – Cold barrier that prevents 9Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing Cold barrier that prevents the reflux of hot air Foto: BSC
  10. 10. Best practice at Marenostrum Problem measured : not all the Racks have the same temperature Proposal: Methacrylate screensMethacrylate screens installed in front of each rack Benefits observed: – All BladeCenters Rack has an l t t / 1ºCequal temperature +/- 1ºC – BladeCenter fan speed reduced 10Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing Foto: BSC
  11. 11. Best practice at Marenostrum Results (including other improvements): reduction of 10% of power consumption (and CO2)consumption (and CO2) Marenostrum power consumption: approx 1 2 Image courtesy of UPC consumption: approx. 1.2 Mwats aprox 1 100 000 €/year 11Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing aprox 1.100.000 €/year
  12. 12. Decreasing DC power consumptionconsumption How can we decrease DC power consumption without a concomitant decrease in the IT load? Can we move the Marenostrum in a better location? e.g. more efficient cooling due to climate conditions 12Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  13. 13. Can we put a DC in a better location?location? I containerized marenostrum and it has traveled with me Helsinki PUE = 1,073 Estimated ~3% of Data courtesy of AST IT Infrastructures (www.AST-global.com) Barcelona PUE = 1,113 13Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing energy could be saved
  14. 14. What else can we do? Maximize work done per watt 14Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  15. 15. What else can we do? Maximize work done per watt Energy consumption is not only determined by the efficiency of the physical resources but it is also dependent on the resource management Intelligent management of resources may lead to i ifi d i f h i bsignificant reduction of the energy consumption by meeting the performance requirements 15Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  16. 16. Contents Motivation for reduction of energy consumption What can we do for DC power efficiency? Power aware Resource Management Research at a Glance Conclusions 16Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  17. 17. Power aware Resource ManagementManagement Different layers inDifferent layers in a Data Center? Data Center Infrastructure Applications System Hardware 17Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  18. 18. Different layers in a Data Center Who is responsible for managing resources? y Who is responsible for managing resources? Data Center Infrastructure … W b S i Applications Web Services Aplication Server Web server System Java Virtual Machine Operating System Hardware Virtualization layer … MIDDLEWARE 18Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing MIDDLEWARE
  19. 19. Power/Energy Aware Resource Management at Middleware levelManagement at Middleware level Challenge: – Shift the paradigm from “time to solution" to "kWh to solution” How?:How?: – Introducing intelligent self-management techniques – Adding energy efficiency parameter to the list of critical operating parameters to control that already including il bilit li bilit favailability, reliability, performance, … 19Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  20. 20. Contents Motivation for reduction of energy consumption What can we do for DC power efficiency? Power aware Resource Management Research at a Glance Conclusions 20Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  21. 21. Reseach at a Glance Today the most common techniques/control knobs of the middleware are related to: – consolidation and virtualization technology, – turning on/off policies, new processors– new processors, – dynamic voltage and frequency scaling, – hybrid architectures on datacenters,y , – federating clouds – etc. 21Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  22. 22. Workload Consolidation Rationalize and reduce operational costs: server consolidation – implies combining workloads from separate applications into a smaller number of systems if there are different peak timessmaller number of systems if there are different peak times C1 rce ement C2 rce ement Time Resour Require Time Resour Require C1 C2 source quirement 22Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing Time Res Req
  23. 23. Virtualization Technologies Consolidation uses virtualization to share the resources – Virtualization software divides a physical server into isolated virtual environments, enabling organizations to run multiple applications or OS on a single serverOS on a single server. Operating System App 3 Operating System App 1 App 2 CPU Mem Virtual Machine CPU Mem Virtual Machine . . . . . . Operating System App 1 App 3App 2 CPU CPU Mem Physical Machine CPU CPUCPU CPU Mem Physical Machine CPU CPU Operating System 23Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing The energy savings can be up to 20% - 70%
  24. 24. Using multi-core processors Middleware allow that applications benefit from new multi-core designs beyond the traditional parallelism Example: Exploitation of Accelerators in memory-boundExample: Exploitation of Accelerators in memory bound Applications using memory compression techniques App 3App 1 App 2 Operating System App 3 Virtual Machine Operating System App 1 App 2 Virtual Machine . . . . . . CPU MemCPU Mem Mem CPU CPU Mem Physical Machine CPU CPUCPU CPU F b d l th i b t 25% 24Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing For memory-bound appl. the energy savings can be up to 25%
  25. 25. Use low-power processors Low-power processors vs high-performance processors Good approach for transactional workloads (webs) – Experiments at BSC: Reduction of 70% 25Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  26. 26. Turn on/off policies However the numerical applications do not behave like this Experiments at BSC with 4 HPC tasks:tasks: • 1 Xeon * 1 hour 317 Watts • 2 Atom * 5 hours 398 Watts – The most energy-efficient approach– The most energy-efficient approach in this environment is to run jobs very fast and then turn off the power of the system 26Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing power of the system.
  27. 27. Hybrid hardwarey But in general the workload in data centers is heterogeneous F HPC X k ffi i tl th At– For HPC, Xeons work more efficiently than Atoms – For Transactional Atoms work more efficiently than XeonsFor Transactional, Atoms work more efficiently than Xeons Therefore, why, y not build the two together?toget e 27Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing Courtesy of Giovanni Leonardi (Mataró-Sicilia)
  28. 28. Hybrid hardwarey Better tradeoff in terms of energy and revenue – Experiments at BSC with heterogeneous workload: Reduction of 25% 28Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  29. 29. In the Cloud Era Resource Management in a Cloud Federationg – Distribution of workload across geographically distributed DC – Consolidation at DC level (and turn off a whole DC) – Possibility to reallocate the workload to a place where energy or cooling is cheaper e.g. solar energy during daytime across different time zones, efficient cooling due to climate conditions etcconditions, etc. – Etc 29Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing Etc.
  30. 30. In the Cloud Era Outsourcing/Insourcing strategiesg g g The energy savings can be up to 10%? 30? 50%? 30Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  31. 31. Conclusions The energy savings can be up to 20% - 70% For memory-bound appl the energy savings can be up to 25% Reduction of 70% Reduction of 25% For memory-bound appl. the energy savings can be up to 25% The energy savings can be up to 10%? 30? 50%? Resource management at Middleware level will play an important role in the reduction of energy consumption Further research is required! 31Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  32. 32. Conclusions Future research directions in resource management – Development of energy-efficient resource management approaches for multi-core CPUs due to the wide adoption of these. – Development of intelligent techniques to manage resources efficienty. – Efficient workload distribution across geographically distributed DC. – SLA negotiation between resource providers and users that includes control over power consumption and/or CO2 emissions.control over power consumption and/or CO2 emissions. – Etc. 32Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  33. 33. Workshop on Green and Environmental Computingp g 27th – 28th October 2010 CSC – IT Center for Science Espoo, Finland Middl hMiddleware research for green Data Centers Prof. Jordi Torres Barcelona Supercomputing Center (BSC) UPC Barcelona TechUPC – Barcelona Tech 33Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing
  34. 34. Workshop on Green and Environmental Computingp g 27th – 28th October 2010 CSC – IT Center for Science Espoo, Finland Jordi Torres has a Master degree in Computer Science from the Technical University of Catalonia (UPC Barcelona Tech, 1988) and also holds a Ph. D. About the author: y ( ) from the same institution (Best UPC Computer Science Thesis Award, 1993). Currently he is a full-time professor in the Computer Architecture Department at UPC and is a manager for the Autonomic Systems andDepartment at UPC and is a manager for the Autonomic Systems and eBusiness Platforms research line at Barcelona Supercomputing Center (BSC). His current principal interest as a researcher is to make IT resources more ffi i t i d t bt i t i bl IT d t f thefficient in order to obtain more sustainable IT and to focus on the resource management needs of modern distributed and parallel cloud computing environments. He has worked on a number of EU and industrial research and development projects. He has more than a hundred publications in the d i l d i l f i th fi ld f di t ib t d darea and was involved in several conferences in the field of distributed and parallel systems. He has been Vice-dean of Institutional Relations at the Computer Science School, and currently he is a member of the University Senate and member 34Jordi Torres – Barcelona Supercomputing Center (BSC) 2010 e-InfraNet Workshop on Green and Environmental Computing , y y of the Board of Governors. He can be reached at www.JordiTorres.org

×