Performance Evaluation of Group Communication Architectures in Large Scale Systems Using MPI

  • Kayhan Erciyes
  • Orhan Dagdeviren
  • Reşat Ümit Payli
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4276)


Group communication is an important paradigm for fault tolerance in large scale systems. We describe various group architectures as pipelined, hierarchical, daisy and hypercube groups each consisting of separate clusters, investigate the theoretical performance bounds of these architectures and evaluate their experimental performances using MPI group communication primitives. We first derive time bounds for multicast message deliveries in these architectures and then provide tests to measure the times taken for the same operation. The multicast message delivery times are tested against the number of clusters within a group and the size of the multicast message. We conclude that daisy architecture is favorable both in terms of delivery times and message sizes theoretically and experimentally.


Group Communication Message Size Hierarchical Architecture Pipeline Architecture Collective Communication 
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.
    Birman, K.P., van Renesse, R.: Reliable Distributed Computing with the Isis Toolkit. IEEE Computer Society Press, Los Alamitos (1994)Google Scholar
  2. 2.
    Chockler, G., Keidar, I., Vitenberg, R.: Group communication specifications: a comprehensive study. ACM Computing Surveys 33(4), 427–469 (2001)CrossRefGoogle Scholar
  3. 3.
    Cristian, F.: Synchronous and Asynchronous Communication. Communications of the ACM. Special Section on Group Communication 39(4) (1996)Google Scholar
  4. 4.
    Amir, Y., et al.: Transis: A communication subsystem for high availability. In: Proc. of 22nd IEEE Int’l. Symp. on Fault-Tolerant Computing, pp. 76–84. IEEE Press, NJGoogle Scholar
  5. 5.
    Van Renesse, R., Birman, K.P., Maffeis, S.: Horus: A Flexible Group communication System. CACM, Special sect. on Group Comm. 39(4) (1996)Google Scholar
  6. 6.
    Amir, Y., et al.: The TOTEM Single Ring Ordering and membership Protocol. ACM Trans. Comp. Systems 13(4) (1995)Google Scholar
  7. 7.
    Keidar, I., et al.: Moshe: A group membership service for WANs. ACM Transactions on Computer Systems (TOCS) 20(3), 191–238 (2002)CrossRefGoogle Scholar
  8. 8.
    Gropp, W., Lusk, E., Doss, N., Skjellum, A.: A High-Performance, Portable Implementation of the MPI Message Passing Interface Standard. In: MPI Developers Conference (1995)Google Scholar
  9. 9.
    Squyres, J.M., Lumsdaine, A., George, W.L., Hagedorn, J.G., Devaney, J.E.: The Interoperable Message Passing Interface (IMPI) Extensions to LAM/MPI. In: MPI Developers Conference, Ithica, NY (2000)Google Scholar
  10. 10.
    Yuan, X., Daniels, S., Faraj, A., Karwande, A.: Group Management Schemes for Implementing MPI Collective Communication over IP Multicast. In: The 6th Int. Conf. on Computer Science and Informatics, Durham, NC, pp. 8–14 (2002)Google Scholar
  11. 11.
    Quinn, M.J.: Parallel Programming in C with MPI and OpenMP, International Edition. Mc Graw Hill (2003)Google Scholar
  12. 12.
    Tunali, T., Erciyes, K., Soysert, Z.: A Hierarchical Fault-Tolerant Ring Protocol For A Distributed Real-Time System. Special issue of Parallel and Distributed Computing Practices on Parallel and Distributed Real-Time Systems 2(1), 33–44 (2000)Google Scholar
  13. 13.
    Allahverdi, N., Kahramanli, S., Erciyes, K.: A Fault Tolerant Routing Algorithm Based on Cube Algebra for Hypercube Systems. JSA 46(2), 201–205 (2000)Google Scholar
  14. 14.
    Grama, A., Gupta, A., Karypis, G., Kumar, V.: Introduction to Parallel Computing, 2nd edn. Addison Wesley Longman, Inc., Amsterdam (2003)Google Scholar
  15. 15.
    Foster, I., Kesselman, C., Tuecke, S.: The Anatomy of the Grid: Enabling Scalable Virtual Organizations. Int. Journal of High Performance Computing Applications 15(3), 200–222 (2001)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Kayhan Erciyes
    • 1
  • Orhan Dagdeviren
    • 1
  • Reşat Ümit Payli
    • 2
  1. 1.Computer Eng. Dept.Izmir Institute of TechnologyUrla, IzmirTurkey
  2. 2.Computational Fluid Dynamics Laboratory, Purdue School of Engineering and TechnologyIndiana University-Purdue UniversityIndianapolis, IndianaU.S.A.

Personalised recommendations