Advertisement

Using Opportunistic combining networks to reduce contention in multicomputers

  • Gregory V. Wilson
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 605)

Abstract

A major difficulty in programming parallel computers is resolving the conflict between centralising data to ensure consistency and distributing data to ensure efficient access. Opportunistic combining networks can reduce access contention without compromising the efficiency of access. The implementation of two simple data structures (a parallel queue and a parallel stack) is described, and performance results obtained by simulation presented.

Keywords

Data Item Tail Index Computing Element Index Server Multistage Interconnection Network 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [BCGL88]
    Robert Bjornson, Nicholas Carriero, David Gelernter, and Jerrold Leichter. Linda, the Portable Parallel. Research Report 520, Department of Computer Science, Yale University, 1988.Google Scholar
  2. [BTK90]
    Henri E. Bal, Andrew S. Tanenbaum, and M. Frans Kaashoek. Orca: A Language for Distributed Programming. ACM SIGPLAN Notices, 25(5), May 1990.Google Scholar
  3. [ChEn91]
    Suresh Chittor and Richard Enbody. Minimizing Contention: A New Mapping Objective for Second-Generation Multicomputer. Technical report, Michigan State University, 1991.Google Scholar
  4. [ClWi91]
    Lyndon J. Clarke and Greg Wilson. Tiny: An Efficient Routing Harness for the Inmos Tranpsuter. Concurrency: Practice and Experience, 3(3), June 1991.Google Scholar
  5. [Fox+88]
    G. Fox, M. Johnson, G. Lyzenga, S. Otto, J. Salmon, and D. Walker. Solving Problems on Concurrent Processors, Volume 1. Prentice-Hall, 1988.Google Scholar
  6. [Got+83]
    Allan Gottlieb, Ralph Grishman, Clyde P. Kruskal, Kevin P. McAuliffe, Larry Rudolph, and Marc Snir. The NYU Ultracomputer — Designing and MIMD Shared Memory Parallel Computer. IEEE Trans. Computers, 32(2), February 1983.Google Scholar
  7. [Harr86]
    Malcolm C. Harrison. The Add-and-Lambda Operation: An Extension of F&A. Technical Report 104, Ultracomputer Research Laboratory, July 1986.Google Scholar
  8. [Harr88]
    Malcolm C. Harrison. Add-and-Lambda II: Eliminating Busy Waits. Technical Report 139, Ultracomputer Research Laboratory, July 1988.Google Scholar
  9. [KuPf86]
    Manoj Kumar and Gregory F. Pfister. The Onset of Hot Spot Contention. In Proc. 1986 Intl. Conf. on Parallel Processing, 1986.Google Scholar
  10. [LKK86]
    Gyungho Lee, Clyde P. Kruskal, and David J. Kuck. The Effectiveness of Combining in Shared Memory Parallel Computers in the Presence of ‘Hot Spots'. In Proc. 1986 Intl. Conf. on Parallel Processing, 1986.Google Scholar
  11. [Pfi+85]
    G.F Pfister, W. C. Brantley, D. A. George, S. L. Harvey, W. J. Kleinfelder, K. P. McAuliffe, E. A. Melton, V. A. Norton, and J. Weiss. The IBM Research Parallel Processor Prototype (RP3): Introduction and Architecture. In Proc. IEEE Intl. Conf. on Parallel Processing, 1985.Google Scholar
  12. [PfNo85]
    G.F Pfister and V. A. Norton. “Hot Spot” Contention and Combining in Multistage Interconnection Networks. In Proc. IEEE Intl. Conf. on Parallel Processing, 1985.Google Scholar
  13. [PrVu81]
    Franco P. Preparata and Jean Vuillemin. The Cube-Connected Cycles: A Versatile Network for Parallel Computation. Comm. ACM, 24(5), May 1981.Google Scholar
  14. [ScBo90]
    Karsten Schwan and Win Bo. Topologies: Distributed Objects on Multicomputers. ACM Trans. on Computer Systems, 8(2):111–157, 1990.CrossRefGoogle Scholar
  15. [Thom86]
    Robert H. Thomas. Behavior of the Butterfly Parallel Processor in the Presence of Memory Hot Spots. In Proc. 1986 Intl. Conf. on Parallel Processing, 1986.Google Scholar
  16. [YTL86]
    Pen-Chung Yew, Nian-Feng Tzeng, and Duncan H. Lawrie. Distibuting Hot-Spot Addressing in Large-Scale Multiprocessors. In Proc. 1986 Intl. Conf. on Parallel Processing, 1986.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Gregory V. Wilson
    • 1
  1. 1.Edinburgh Parallel Computing Centre James Clerk Maxwell BuildingUniversity of EdinburghEdinburgh

Personalised recommendations