Advertisement

The Journal of Supercomputing

, Volume 73, Issue 6, pp 2369–2401 | Cite as

Energy consumption reduction for asynchronous message-passing applications

  • Ahmed Fanfakh
  • Jean-Claude CharrEmail author
  • Raphaël Couturier
  • Arnaud Giersch
Article

Abstract

It is widely accepted that the asynchronous parallel methods are more suitable than the synchronous ones on a grid architecture. Indeed, they outperform the synchronous methods, because they overlap the communications of the synchronous methods with computations. However, they also usually execute more iterations than the synchronous ones and thus consume more energy. To reduce the energy consumption of the CPUs executing such methods, the Dynamic voltage and frequency scaling technique can be used. It lowers the frequency of a CPU to reduce its energy consumption, but it also decreases its computing power. Therefore, the frequency that gives the best trade-off between energy consumption and performance must be selected. This paper presents a new online frequency selecting algorithm for parallel iterative asynchronous methods running over grids. It selects a vector of frequencies that gives the best trade-off between energy consumption and performance. New energy and performance models were used in this algorithm to predict the execution time and the energy consumption of synchronous, asynchronous, or hybrid iterative applications running over grids. The proposed algorithm was evaluated on the SimGrid simulator. The experiments showed that synchronously applying the proposed algorithm to the asynchronous version of the application reduces on average its energy consumption by 22% and speeds it up by 5.72%. Finally, the proposed algorithm was also compared to a method that uses the well-known energy and delay product and the comparison results showed that it outperforms this method in terms of energy consumption and performance.

Keywords

Energy optimization DVFS Asynchronous computing Grid computing 

Notes

Acknowledgements

This work has been partially supported by the Labex ACTION project (contract “ANR-11-LABX-01-01”). Computations have been performed on the supercomputer facilities of the Mésocentre de calcul de Franche-Comté. As a Ph.D. student, Mr. Ahmed Fanfakh, would like to thank the University of Babylon (Iraq) for supporting his work.

References

  1. 1.
    Anzt H (2012) Asynchronous and multiprecision linear solvers, Ph.D. thesis. Karlsruher Institut für Technologie, Bade-WurtembergGoogle Scholar
  2. 2.
    Bahi J, Contassot-Vivier S, Couturier R (2007) Parallel iterative algorithms: from sequential to grid computing, Numerical Analysis and Scientific Computating, vol 1. Chapman and Hall/CRCGoogle Scholar
  3. 3.
    Baldassin A, de Carvalho J, Garcia L, Azevedo R (2012) Energy-performance tradeoffs in software transactional memory. In: 2012 IEEE 24th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), pp 147–154Google Scholar
  4. 4.
    Beloglazov A, Buyya R (2012) Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in cloud data centers. Concurr Comput Pract Exp 24(13):1397–1420CrossRefGoogle Scholar
  5. 5.
    Casanova H, Giersch A, Legrand A, Quinson M, Suter F (2014) Versatile, scalable, and accurate simulation of distributed applications and platforms. J Parallel Distrib Comput 74(10):2899–2917CrossRefGoogle Scholar
  6. 6.
    Charr JC, Couturier R, Fanfakh A, Giersch A (2014) Dynamic frequency scaling for energy consumption reduction in distributed MPI programs. In: ISPA 2014: 12th IEEE International Symposium on Parallel and Distributed Processing with Applications. IEEE Computer Society, Milan, pp 225–230Google Scholar
  7. 7.
    Charr JC, Couturier R, Fanfakh A, Giersch A (2015) Energy consumption reduction with DVFS for message passing iterative applications on heterogeneous architectures. In: The 16th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing. IEEE Computer Society, IndiaGoogle Scholar
  8. 8.
    Chen JJ, Huang K, Thiele L (2012) Dynamic frequency scaling schemes for heterogeneous clusters under quality of service requirements. Inf Sci Eng 28(6):1073–1090Google Scholar
  9. 9.
    Cocana-Fernandez A, Sanchez L, Ranilla J (2015) A software tool to efficiently manage the energy consumption of hpc clusters. In: 2015 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), pp 1–8Google Scholar
  10. 10.
    Cochran R, Hankendi C, Coskun A, Reda S (2011) Identifying the optimal energy-efficient operating points of parallel workloads. In: Proceedings of the International Conference on Computer-Aided Design. IEEE Press, NJ, pp 608–615Google Scholar
  11. 11.
    Da Costa G, de Assunção MD, Gelas JP, Georgiou Y, Lefèvre L, Orgerie AC, Pierson JM, Richard O, Sayah A (2010) Multi-facet approach to reduce energy consumption in clouds and grids: the green-net framework. In: Proceedings of the 1st International Conference on Energy-Efficient Computing and Networking. ACM, New York, pp 95–104Google Scholar
  12. 12.
    Da Costa G, Gelas JP, Georgiou Y, Lefevre L, Orgerie AC, Pierson J, Richard O, Sharma K (2009) The green-net framework: energy efficiency in large scale distributed systems. In: IEEE International Symposium on Parallel Distributed Processing. IPDPS 2009, pp 1–8Google Scholar
  13. 13.
    Freeh VW, Pan F, Kappiah N, Lowenthal DK, Springer R (2005) Exploring the energy-time tradeoff in MPI programs on a power-scalable cluster. In: Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium, pp 4a–4a. IEEE Computer Society, Washington, DCGoogle Scholar
  14. 14.
    Ge R, Vogt R, Majumder J, Alam A, Burtscher M, Zong Z (2013) Effects of dynamic voltage and frequency scaling on a k20 gpu. In: 2013 42nd International Conference on Parallel Processing (ICPP), pp 826–833Google Scholar
  15. 15.
    Guermouche A, Triquenaux N, Pradelle B, Jalby W (2015) Minimizing energy consumption of MPI programs in realistic environment. Computing Research RepositoryGoogle Scholar
  16. 16.
    Guérout T, Monteil T, Costa GD, Calheiros RN, Buyya R, Alexandru M (2013) Energy-aware simulation with DVFS. Simul Model Pract Theory 39:76–91CrossRefGoogle Scholar
  17. 17.
    Hsu CH, Chun Feng W (2005) A power-aware run-time system for high-performance computing. In: Proceedings of the ACM/IEEE SC 2005 Conference on Supercomputing, 2005, pp 1–9Google Scholar
  18. 18.
    Joshi KR, Hiltunen MA, Schlichting RD, Sanders WH (2010) Blackbox prediction of the impact of DVFS on end-to-end performance of multitier systems. ACM SIGMETRICS Perform Eval Rev 37(4):59–63CrossRefGoogle Scholar
  19. 19.
    Kim NS, Austin T, Blaauw D, Mudge T, Flautner K, Hu JS, Irwin MJ, Kandemir M, Narayanan V (2003) Leakage current: Moore’s law meets static power 36(12):68–75Google Scholar
  20. 20.
    Le Sueur E, Heiser G (2010) Dynamic voltage and frequency scaling: The laws of diminishing returns. In: Proceedings of the 2010 Workshop on Power Aware Computing and Systems (HotPower’10)Google Scholar
  21. 21.
    Ma K, Li X, Chen W, Zhang C, Wang X (2012) Greengpu: a holistic approach to energy efficiency in gpu–cpu heterogeneous architectures. In: 2012 41st International Conference on Parallel Processing (ICPP), pp 48–57Google Scholar
  22. 22.
    Malkowski K (2009) Co-adapting scientific applications and architectures toward energy-efficient high performance computing, Ph.D. thesis. The Pennsylvania State University, USAGoogle Scholar
  23. 23.
    Muralimanohar N, Ramani K, Balasubramonian R (2006) Power efficient resource scaling in partitioned architectures through dynamic heterogeneity. In: Proceedings of ISPASSGoogle Scholar
  24. 24.
    O’Leary DP, White RE (1985) Multi-splittings of matrices and parallel solution of linear systems. SIAM J Algebr Discrete Methods 6(4):630–640CrossRefzbMATHGoogle Scholar
  25. 25.
    Orgerie AC, Lefevre L, Gelas JP (2008) Save watts in your grid: Green strategies for energy-aware framework in large scale distributed systems. In: 14th IEEE International Conference on Parallel and Distributed Systems, 2008. ICPADS ’08, pp 171–178Google Scholar
  26. 26.
    Peraza J, Tiwari A, Laurenzano ML, Snavely C (2012) PMaC’s green queue: a framework for selecting energy optimal DVFS configurations in large scale MPI applications. Concurrency Computation: Practice and Experience, pp 1–20Google Scholar
  27. 27.
    Pietri I, Sakellariou R (2014) Energy-aware workflow scheduling using frequency scaling. In: 2014 43rd International Conference on Parallel Processing Workshops (ICCPW), pp 104–113Google Scholar
  28. 28.
    Ramamonjisoa C, Ziane Khodja L, Laiymani D, Giersch A, Couturier R (2014) Simulation of asynchronous iterative algorithms using simgrid. In: 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS) High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, pp 890–895Google Scholar
  29. 29.
    Rauber T, Rünger G, Schwind M, Xu H, Melzner S (2014) Energy measurement, modeling, and prediction for processors with frequency scaling. J Supercomput 70(3):1451–1476CrossRefGoogle Scholar
  30. 30.
    Rauber T, Rünger G (2012) Analytical modeling and simulation of the energy consumption of independent tasks. In: Proceedings of the Winter Simulation Conference. Winter Simulation Conference, pp 245:1–245:13Google Scholar
  31. 31.
    Rizvandi NB, Taheri J, Zomaya AY (2011) Some observations on optimal frequency selection in DVFS-based energy consumption minimization. J Parallel Distrib Comput 71(8):1154–1164CrossRefzbMATHGoogle Scholar
  32. 32.
    Shelepov D, Fedorova A (2008) Scheduling on heterogeneous multicore processors using architectural signatures. In: Workshop on the Interaction between Operating Systems and Computer Architecture, in conjunction with ISCAGoogle Scholar
  33. 33.
    Shen H, Lu J, Qiu Q (2012) Learning based DVFS for simultaneous temperature, performance and energy management. In: ISQED, pp 747–754Google Scholar
  34. 34.
    Spiliopoulos V, Kaxiras S, Keramidas G (2011) Green governors: a framework for continuously adaptive dvfs. In: International Green Computing Conference and Workshops (IGCC), pp 1–8Google Scholar
  35. 35.
    Thiam C, Da Costa G, Pierson JM (2014) Energy aware clouds scheduling using anti-load balancing algorithm : EACAB. In: 3rd International Conference on Smart Grids and Green IT Systems (SMARTGREENS 2014), Barcelona, pp 82–89Google Scholar
  36. 36.
    Vishnu A, Song S, Marquez A, Barker K, Kerbyson D, Cameron K, Balaji P (2013) Designing energy efficient communication runtime systems: a view from pgas models. J Supercomput 63(3):691–709CrossRefGoogle Scholar
  37. 37.
    Wang L, Khan SU, Chen D, Kołodziej J, Ranjan R, zhong Xu C, Zomaya A (2013) Energy-aware parallel task scheduling in a cluster. Future Gen Comput Syst 29(7):1661–1670CrossRefGoogle Scholar
  38. 38.
    Liang W-L, Lai P-T, Chiou CW (2010) An energy conservation dvfs algorithm for the android operating system. J Converg 1(1):93–100Google Scholar
  39. 39.
    Zapater M, Ayala JL, Moya JM, Vaidyanathan K, Gross K, Coskun AK (2013) Leakage and temperature aware server control for improving energy efficiency in data centers. In: Proceedings of the Conference on Design, Automation and Test in Europe, San Jose, pp 266–269Google Scholar
  40. 40.
    Zapater M, Tuncer O, Ayala J, Moya J, Vaidyanathan K, Gross K, Coskun A (2015) Leakage-aware cooling management for improving server energy efficiency. IEEE Trans Parallel Distrib Syst 26(10):2764–2777CrossRefGoogle Scholar
  41. 41.
    Zhu Y, Mueller F (2007) Exploiting synchronous and asynchronous DVS for feedback EDF scheduling on an embedded platform. ACM Trans Embed Comput 7(1)Google Scholar
  42. 42.
    Zhuo J, Chakrabarti C (2008) Energy-efficient dynamic task scheduling algorithms for dvs systems. ACM Trans Embed Comput 7(2):17:1–17:25Google Scholar
  43. 43.
    Zong Z, Manzanares A, Ruan X, Qin X (2011) Ead and pebd: two energy-aware duplication scheduling algorithms for parallel tasks on homogeneous clusters. IEEE Trans Comput 60(3):360–374MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Ahmed Fanfakh
    • 1
  • Jean-Claude Charr
    • 1
    Email author
  • Raphaël Couturier
    • 1
  • Arnaud Giersch
    • 1
  1. 1.FEMTO-ST Institute, CNRS, Univ. Bourgogne Franche-Comté (UBFC), IUT de Belfort-MontbéliardBelfortFrance

Personalised recommendations