Modeling Message-Passing Overhead on NCHC Formosa PC Cluster

  • Chau-Yi Chou
  • Hsi-Ya Chang
  • Shuen-Tai Wang
  • Shou-Cheng Tcheng
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3947)


The communication plays a role in the overall system performance. The characterization of the communication overhead is very important to estimate the global performance of parallel applications and to detect possible bottlenecks. In this work, we evaluate and model the performance of the message-passing libraries on NCHC Formosa PC Cluster, a large cluster system with dual processor nodes and connected by Gigabits Ethernet networks. Our aim is to fairly characterize the communication primitives using general models and performance metrics. We use the formulae to estimate the communication time of a real application program for molecular dynamics simulation. We hope that it is able to provide some useful information for performance prediction and scientific computing.


Communication Overhead Communication Time Message Size Message Length Dedicated Node 
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.
    Sterling, T., Becker, D., Savarese, D., et al.: BEOWULF: A Parallel Workstation for Scientific Computation. In: Proc. of the 1995 International Conf. on Parallel Processing (1995)Google Scholar
  2. 2.
    Sterling, T., Savarese, D., Becker, D., et al.: Communication Overhead for Space Science Applications on the Beowulf Parallel Workstation. In: Proc. of 4th IEEE Symposium on High Performance Distributed Computing (1995)Google Scholar
  3. 3.
    Reschke, C., Sterling, T., Ridge, D.: A Design Study of Alternative Network Topologies for the Beowulf Parallel Workstation. In: Proceedings of the 5th IEEE Symposium on High Performance and Distributed Computing (1996)Google Scholar
  4. 4.
    Ridge, D., Becker, D., Merkey, P.: Beowulf: Harnessing the Power of Parallelism in a Pile-of-PCs. In: Proceedings of IEEE Aerospace (1997)Google Scholar
  5. 5.
    Pfister, G.F.: In Search of Clusters. Prentice-Hall, Inc., Englewood Cliffs (1998)Google Scholar
  6. 6.
    Xu, Z., Hwang, K.: Modeling Communication Overhead: MPI and MPL Performance on the IBM SP2. IEEE Parallel & Distributed Technology 4(1), 9–23 (1996)CrossRefGoogle Scholar
  7. 7.
    Prieto, M., Espadas, D., Llorente, I.M., Tirado, F.: Message Passing Evaluation and Analysis on Cray T3E and SGI Origin 2000 Systems. In: Amestoy, P.R., Berger, P., Daydé, M., Duff, I.S., Frayssé, V., Giraud, L., Ruiz, D. (eds.) Euro-Par 1999. LNCS, vol. 1685, pp. 173–182. Springer, Heidelberg (1999)CrossRefGoogle Scholar
  8. 8.
    Touriño, J., Doallo, R.: Characterization of Message-Passing Overhead on the AP3000 Multicomputer. In: International Conference on Parallel Processing (2001)Google Scholar
  9. 9.
    Gunawan, T., Cai, W.: Performance Analysis of a Myrinet-Based Cluster. Cluster Computing 6, 229–313 (2003)CrossRefGoogle Scholar
  10. 10.
    NCHC Formosa PC Cluster Home Page,
  11. 11.
    Top 500 List,
  12. 12.
    Hockney, R.W.: Performance Parameters and benchmarking of supercomputers. Parallel Computing 17, 1111–1130 (1991)CrossRefGoogle Scholar
  13. 13.
    Hockney, R.W.: The Communication Challenge for MPP: Intel Paragon and Meiko CS-2. Parallel Computing 20, 389–398 (1994)CrossRefGoogle Scholar
  14. 14.
    Burns, G., Daoud, R., Vaigl, J.: LAM: An Open Cluster Environment for MPI. In: Proceedings of Supercomputing Symposium 1994, pp. 379–386 (1994)Google Scholar
  15. 15.
    Lichten, W.: Data and Error Analysis. Prentice Hall, Englewood Cliffs (1998)MATHGoogle Scholar
  16. 16.
    Huang, K.-C., Chang, H.-Y., Shen, C.-Y., Chou, C.-Y., Tcheng, S.-C.: Benchmarking and Performance Evaluation of NCHC PC Cluster. In: High Performance Computing in the Asia-Pacific Region (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Chau-Yi Chou
    • 1
  • Hsi-Ya Chang
    • 1
  • Shuen-Tai Wang
    • 1
  • Shou-Cheng Tcheng
    • 1
  1. 1.National Center For High-Performance ComputingChina

Personalised recommendations