The Journal of Supercomputing

, Volume 71, Issue 3, pp 1163–1174 | Cite as

Energy-efficient allocation of computing node slots in HPC clusters through parameter learning and hybrid genetic fuzzy system modeling

  • Alberto Cocaña-Fernández
  • Jose Ranilla
  • Luciano Sánchez
Article
First Online:

Abstract

Decision-making mechanisms for online allocation of computer node slots in HPC clusters are commonly based on simple knowledge-based systems comprised of individual sets of if–then rules. In contrast with previous works where these rules were designed using expert knowledge, two different types of evolutionary learning algorithms are compared in this paper. In the first case, some of the numerical parameters defining a human-designed knowledge base are tuned. In the second case, a genetic fuzzy system evolves a partial rule set that, after being combined with some expert rules, conforms the most appropriate knowledge base for a given load scenario. In both cases, the proposed approaches optimize the quality of service and the number of node reconfigurations along with the energy consumption. An experimental study has been made using actual workloads from the Scientific Modeling Cluster at Oviedo University, and statistical evidence was found supporting the adoption of the new learning system.

Keywords

Energy-efficient cluster computing Multi-criteria decision making Evolutionary algorithms 
This is a preview of subscription content, log in to check access.

Notes

Acknowledgments

This work has been partially supported by “Ministerio de Economía y Competitividad” from Spain/FEDER under grants TEC2012-38142-C04-04 and TIN2011-24302.

References

  1. 1.
    Alonso P, Badia RM, Labarta J, Barreda M, Dolz MF, Mayo R, Quintana-Orti ES, Reyes R (2012) Tools for power-energy modelling and analysis of parallel scientific applications. In: 2012 41st international conference on parallel processing. IEEE, New Jersey, pp 420–429Google Scholar
  2. 2.
    Alvarruiz F, de Alfonso C, Caballer M, Hernández V (2012) An energy manager for high performance computer clusters. In: 2012 IEEE 10th international symposium on parallel and distributed processing with applications. IEEE, New Jersey, pp 231–238Google Scholar
  3. 3.
    Bash C, Forman G (2007) Cool job allocation: measuring the power savings of placing jobs at cooling-efficient locations in the data center. USENIX Association, Berkeley, p 29Google Scholar
  4. 4.
    Berral JL, Goiri Í, Nou R, Julià F, Guitart J, Gavaldà R, Torres J (2010) Towards energy-aware scheduling in data centers using machine learning. In: Proceedings of the 1st international conference on energy-efficient computing and networking—e-energy ’10. ACM Press, New York, p 215Google Scholar
  5. 5.
    Buyya R, Jin H, Cortes R (2002) Cluster computing. Future Gener Comput Syst 18(3):v–viiiGoogle Scholar
  6. 6.
    Cheng Y, Zeng Y (2011) Automatic energy status controlling with dynamic voltage scaling in power-aware high performance computing cluster. In: 2011 12th international conference on parallel and distributed computing, applications and technologies. IEEE, New York, pp 412–416Google Scholar
  7. 7.
    Chetsa GLT, Lefrvre L, Pierson JM, Stolf P, Da Costa G (2012) A runtime framework for energy efficient HPC systems without a priori knowledge of applications. In: 2012 IEEE 18th international conference on parallel and distributed systems. IEEE, New York, pp 660–667Google Scholar
  8. 8.
    Cocaña Fernández A, Ranilla J, Sánchez L (2014) Energy-efficient allocation of computing node slots in hpc clusters through evolutionary multi-criteria decision making. In: Proceedings of the 14th international conference on computational and mathematical methods in science and engineering, CMMSE 2014, pp 318–330Google Scholar
  9. 9.
    Das R, Kephart JO, Lefurgy C, Tesauro G, Levine DW, Chan H (2008) Autonomic multi-agent management of power and performance in data centers, pp 107–114Google Scholar
  10. 10.
    Deb K, Pratap A, Agarwal S, Meyarivan T (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 6(2):182–197CrossRefGoogle Scholar
  11. 11.
    Dolz MF, Fernández JC, Iserte S, Mayo R, Quintana-Ortí ES, Cotallo ME, Díaz G (2011) Energy saving cluster experience in CETA-CIEMAT. In: 5th Iberian GRID infrastructure conference, SantanderGoogle Scholar
  12. 12.
    Matthew E, Mike A, Felipe FC, da Fonseca M, Para GV, Michael S (2011) Smarter data centers: achieving greater efficiency. Technical report, IBM RedpaperGoogle Scholar
  13. 13.
    EIA. Electric Power Monthly—Energy Information AdministrationGoogle Scholar
  14. 14.
    Elnozahy EN, Kistler M, Rajamony R (2002) Energy-efficient server clusters, pp 179–197Google Scholar
  15. 15.
    Emerson Network Power (2009) Energy logic: reducing data center energy consumption by creating savings that cascade across systems. Technical reportGoogle Scholar
  16. 16.
    Eurostat (2013) Electricity and natural gas price statistics—statistics explainedGoogle Scholar
  17. 17.
    Freeh VW, Lowenthal DK (2005) Using multiple energy gears in MPI programs on a power-scalable cluster. In: Proceedings of the tenth ACM SIGPLAN symposium on principles and practice of parallel programming—PPoPP ’05. ACM Press, New York, p 164Google Scholar
  18. 18.
    Freeh VW, Lowenthal DK, Pan F, Kappiah N, Springer R, Rountree BL, Femal ME (2007) Analyzing the energy–time trade-off in high-performance computing applications. IEEE Trans Parallel Distrib Syst 18(6):835–848CrossRefGoogle Scholar
  19. 19.
    Garcia DF, Entrialgo J, Garcia J, Garcia M (2010) A self-managing strategy for balancing response time and power consumption in heterogeneous server clusters. In: 2010 international conference on electronics and information engineering, vol 1. IEEE, New York, pp V1–537-V1-541Google Scholar
  20. 20.
    Gartner (2007) Gartner estimates ICT industry accounts for 2 percent of global CO2 emissionsGoogle Scholar
  21. 21.
    Ge R, Feng X, Feng W, Cameron KW (2007) CPU MISER: a performance-directed, run-time system for power-aware clusters. In: 2007 international conference on parallel processing (ICPP 2007). IEEE, New York, pp 18–18Google Scholar
  22. 22.
    Ruud H (2011) The blue gene/Q compute chip. Technical report, IBM CorporationGoogle Scholar
  23. 23.
    Hsu CH, Feng W (2005) A power-aware run-time system for high-performance computing. In: ACM/IEEE SC 2005 conference (SC’05). IEEE, New York, pp 1–1Google Scholar
  24. 24.
    Hsu CH, Kremer U (2003) The design, implementation, and evaluation of a compiler algorithm for CPU energy reduction. ACM SIGPLAN Not 38(5):38CrossRefGoogle Scholar
  25. 25.
    Huang S, Feng W (2009) Energy-efficient cluster computing via accurate workload characterization. In: 2009 9th IEEE/ACM international symposium on cluster computing and the grid. IEEE, New York, pp 68–75Google Scholar
  26. 26.
    IBM Systems and Technology Group (2011) IBM system blue gene/Q—DCD12345USEN.pdf. Technical report, IBM, Somers, NYGoogle Scholar
  27. 27.
    Ishibuchi H, Nakashima T, Nii M (2004) Classification and modeling with linguistic information granules: advanced approaches to linguistic data mining. Adv Inf ProcessGoogle Scholar
  28. 28.
    Lang W, Patel JM, Naughton JF (2010) On energy management, load balancing and replication. ACM SIGMOD Record 38(4):35CrossRefGoogle Scholar
  29. 29.
    Li D, Nikolopoulos DS, Cameron K, de Supinski BR, Schulz M (2010) Power-aware MPI task aggregation prediction for high-end computing systems. In: 2010 IEEE international symposium on parallel & distributed processing (IPDPS). IEEE, New York, pp 1–12Google Scholar
  30. 30.
    Lim M, Freeh V, Lowenthal D (2006) Adaptive, transparent frequency and voltage scaling of communication phases in MPI programs. In: ACM/IEEE SC 2006 conference (SC’06). IEEE, New York, p 14Google Scholar
  31. 31.
    Llamas RM, Garcia DF, Entrialgo J (2012) A technique for self-optimizing scalable and dependable server clusters under QoS constraints. In: 2012 IEEE 11th international symposium on network computing and applications. IEEE, New York, pp 61–66Google Scholar
  32. 32.
    Pinheiro E, Bianchini R, Carrera EV, Heath T (2001) Load balancing and unbalancing for power and performance in cluster-based systems. In: Workshop on compilers and operating systems for low power, vol 180. Barcelona, Spain, pp 182–195Google Scholar
  33. 33.
    Schubert S, Kostic D, Zwaenepoel W, Shin KG (2012) Profiling software for energy consumption. In: 2012 IEEE international conference on green computing and communications. IEEE, New York, pp 515–522Google Scholar
  34. 34.
    Takagi T, Sugeno M (1985) Fuzzy identification of systems and its applications to modeling and control. IEEE Trans Syst Man Cybern, SMC-15(1):116–132Google Scholar
  35. 35.
    Tang G, Gupta Q, Varsamopoulos SKS (2008) Energy-efficient thermal-aware task scheduling for homogeneous high-performance computing data centers: a cyber-physical approach. IEEE Trans Parallel Distrib Syst 19(11):1458–1472Google Scholar
  36. 36.
    Unit The Economist Intelligence (2007) IT and the environment A new item on the CIOGs agenda? Technical report, The EconomistGoogle Scholar
  37. 37.
    U.S. Environmental Protection Agency (2007) Report to congress on server and data center energy efficiency public law. Technical report, ENERGY STAR Program, pp 109–431Google Scholar
  38. 38.
    Lassonde W, Khan SU, Min-Allah N, Madani SA, Li J, Zhang L, Wang L, Ghani N, Kolodziej J, Li H, Zomaya AY, Xu CZ, Balaji P, Vishnu A, Pinel F, Pecero JE, Kliazovich D, Bouvry P (2011) An overview of energy efficiency techniques in cluster computing systems. Cluster Comput 16(1):3–15Google Scholar
  39. 39.
    Xian C, Lu YH, Li Z (2007) A programming environment with runtime energy characterization for energy-aware applications. In: Proceedings of the 2007 international symposium on low power electronics and design—ISLPED ’07. ACM Press, New York, pp 141–146Google Scholar
  40. 40.
    Xue Z, Dong X, Ma S, Fan S, Mei Y (2007) An energy-efficient management mechanism for large-scale server clusters. In: The 2nd IEEE Asia-Pacific service computing conference (APSCC 2007). IEEE, New York, pp 509–516Google Scholar
  41. 41.
    Yeo CS, Buyya R, Pourreza H, Rasit Eskicioglu M, Graham P, Pourreza P, Sommers F (2006) Cluster computing: high-performance, high-availability, and high-throughput processing on a network of computers. In: Zomaya AY (ed) Handbook of nature-inspired and innovative computing. Springer, Berlin, pp 521–551Google Scholar
  42. 42.
    Zong Z, Nijim M, Manzanares A, Qin X (2007) Energy efficient scheduling for parallel applications on mobile clusters. Cluster Comput 11(1):91–113CrossRefGoogle Scholar
  43. 43.
    Zong Z, Ruan X, Manzanares A, Bellam K, Qin X (2010) Improving energy-efficiency of computational grids via scheduling. In: Antonopoulos N, Exarchakos G, Li M, Liotta A (eds) Handbook of research on P2P and grid systems for service-oriented computing, chap. 22. IGI Global, HersheyGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Alberto Cocaña-Fernández
    • 1
  • Jose Ranilla
    • 1
  • Luciano Sánchez
    • 1
  1. 1.Departamento de InformáticaUniversidad de OviedoGijónSpain

Personalised recommendations