Compiler-Directed Energy-Time Tradeoff in MPI Programs on DVS-Enabled Parallel Systems

  • Huizhan Yi
  • Juan Chen
  • Xunjun Yang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4330)


Although parallel systems with high peak performance have been exciting, high peak performance often means high power consumption. In this paper, power-aware parallel systems are investigated, where each node can make dynamic voltage scaling (DVS). Based on the characteristics of communication and memory access in MPI programs, a compiler is used to automatically form communication and computation regions, and to optimally assign frequency and voltage to the regions. Frequency and voltage of each node are dynamically adjusted, and energy consumption is minimized within the limit of performance loss. The results from simulations and experiments show that compiler-directed energy-time tradeoff can save 20~40% energy consumption with less than 5% performance loss.


Execution Time Performance Loss Communication Function Parallel Application Computation Region 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Feng, X., Ge, R., Cameron, K.W.: Power and Energy Profiling of Scientific Applications on Distributed Systems. In: 19th International Parallel and Distributed Processing Symposium (IPDPS 2005), Denver, CO (April 2005)Google Scholar
  2. 2.
    The BlueGene/L Team. An Overview of the BlueGene/L Supercomputer. In: SC 2002, November 16-22, Baltimore, USA (2002) Google Scholar
  3. 3.
    Hsu, C., Feng, W.: A Power-Aware Run-Time System for High-Performance Computing. In: SC 2005, November 12-18, Seattle, Washington, USA (2005)Google Scholar
  4. 4.
    Pering, T., Burd, T., Brodersen, R.: Dynamic Voltage Scaling and the Design of a Low-Power Microprocessor System. In: Proc. Power-Driven Microarchitecture Workshop, associated with ISCA 1998, Barcelona, Spain (June 1998)Google Scholar
  5. 5.
    Ge, R., Feng, X., Cameron, K.W.: Performance-constrained Distributed DVS Scheduling for Scientific Applications on Power-aware Clusters. In: SC 2005, November 12-18, Seattle, Washington, USA (2005)Google Scholar
  6. 6.
    Freeh, V.W., Pan, F., Kappiah, N., Lowenthal, D.K., Springer, R.: Exploring the Energy-Time Tradeoff in MPI Programs on a Power-Scalable Cluster. In: 19th International Parallel and Distributed Processing Symposium (IPDPS 2005), Denver, CO (April 2005)Google Scholar
  7. 7.
    Weiser, W., Welch, B., Demers, A., Shenker, S.: Scheduling for Reduced CPU Energy. In: Proceedings of the 1st USENIX Symposium on Operating Systems Design and Implementation, November 1994, pp. 13–23 (1994)Google Scholar
  8. 8.
    Lorch, J.R.: Operating Systems Techniques for Reducing Processor Energy Consumption [Ph.D. thesis]. UNIVERSITY of CALIFORNIA, BERKELEY (Fall, 2001)Google Scholar
  9. 9.
    Mosse, D., et al.: Compiler-Assisted Dynamic Power-Aware Scheduling for Real-Time Applications. In: Workshop on Compilers and Operating Systems for Low-Power (COLP 2000), Philadelphia, PA (October 2000)Google Scholar
  10. 10.
    Hsu, C.-H., et al.: The Design, Implementation, and Evaluation of a Compiler Algorithm for CPU Energy Reduction. In: Proceedings of the ACM SIGPLAN 2003 conference on Programming language design and implementation, pp. 38–48 (June 2003)Google Scholar
  11. 11.
    Springer, R., Lowenthal, D.K., Rountree, B., Freeh, V.W.: Minimizing Execution Time in MPI Programs on an Energy-Constrained, Power-Scalable Cluster. In: PPoPP 2006, New York, March 29–31 (2006)Google Scholar
  12. 12.
    Wakabayashi, M., Amano, H.: Envionment for Multiprocessor Simulator Development. In: Fifth International Symposium on Parallel Architectures, Algorithms, and Networks, December 2000, pp. 64–71 (2000)Google Scholar
  13. 13.
    Hsu, C.-H.: Compiler-Directed Dynamic Voltage and Frequency Scaling for CPU Power and Energy Reduction [Ph.D, thesis], The State University of New Jersey (October 2003)Google Scholar
  14. 14.
    Burd, T.D., Brodersen, R.W.: Design Issues for Dynamic Voltage Scaling. In: ISPLED 2000, Rapallo, Italy (2000)Google Scholar
  15. 15.
    Radhakrishnan, R., Pankratz, J.: Introducing DDR2 Memory in Eighth-Generation Dell PowerEdge Servers for Improved Performance. Dell Power Solutions (October 2004)Google Scholar
  16. 16.
    Voltaire Inc., InfiniBand: The Next Step in High Performance Computing, A Voltaire White Paper (February 2003)Google Scholar
  17. 17.
    Thakur, R., Rabenseifnerm, R., Gropp, Q.: Optimization of Collective Communication Operations in MPICH. International Journal of High Performance Computing Applications (2005)Google Scholar
  18. 18.
    Thies, W.: StreamIt: A Compiler Infrastructure for Stream Programs. In: International Conference on Compiler Construction, Grenoble, France (April 2002)Google Scholar
  19. 19.
    Saavedra-Barrera, R.H.: Machine Characterization and Benchmark Performance Prediction, Berkeley, California, June 30 (1988)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Huizhan Yi
    • 1
  • Juan Chen
    • 1
  • Xunjun Yang
    • 1
  1. 1.Section 620, School of ComputerNational University of Defense TechnologyChangshaP.R. China

Personalised recommendations