Skip to main content
SpringerLink
Log in

Routing with finite speeds of memory and network

  • Contributed Papers
  • Conference paper
  • First Online:
Mathematical Foundations of Computer Science 1997 (MFCS 1997)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 1295))

Abstract

On practical parallel computers, the time for routing a distribution of sufficiently large packets can be approximated by max{T f ,T b }. Here T f is proportional to the maximum number of bytes a PU sends and receives, and Tb is proportional to the maximum number of bytes a connection in the network has to transfer. We show that several important routing patterns can be performed by a sequence of balanced all-to-all routings and analyze how to optimally perform these under the above cost-model. We concentrate on dimension-order routing on meshes, and assume that the routing pattern must be decomposed into a sequence of permutations. The developed strategy has been implemented on the Intel Paragon. In comparison with the trivial strategy, in which PU; routes to PU (i+t)mod P to permutation t, 1 ≤ t < P, one gains between 10 and 20%.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Ding, K.-S., C.-T. Ho, J. J. Tsay, ‘Matrix Transpose on Meshes with Wormhole and XY Routing,’ Proc. 6th Symposium on Parallel and Distributed Processing, pp. 656–663, IEEE, 1994.

    Google Scholar 

  2. Hambrusch, S.E., F. Hameed, A. Khokhar, ‘Communication Operations on Coarse-Grained Architectures,’ Parallel Computing, 21, pp. 731–751, 1995.

    Article  Google Scholar 

  3. Kaufmann, M., U. Meyer, J.F. Sibeyn, ‘Matrix Transpose on Meshes: Theory and Practice,’ Proc. 11th International Parallel Processing Symposium, IEEE, 1997, to appear.

    Google Scholar 

  4. Scott, D.S., ‘Efficient All-to-All Communication Patterns in Hypercube and Mesh Topologies,’ Proc. 6th Distributed Memory Computing Conference, pp. 398–403, 1991.

    Google Scholar 

  5. Sibeyn, J.F., F. Guillaume, T. Seidel, ‘Practical Parallel List Ranking,’ Proc. 4th Symposium on Solving Irregularly Structured Problems in Parallel, LNCS, Springer-Verlag, 1997. To appear.

    Google Scholar 

  6. Stricker, T.M., J.C. Hardwick, ‘From AAPC Algorithms to High Performance Permutation Routing and Sorting,’ Proc. 8th Symp. on Parallel Algorithms and Architectures, pp. 200–203, ACM, 1996.

    Google Scholar 

  7. Valiant, L.G., G.J. Brebner, ‘Universal Schemes for Parallel Communication,’ Proc. 13th Symposium on Theory of Computing, pp. 263–277, ACM, 1981.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Max-Planck-Institut für Informatik, Im Stadtwald, 66123, Saarbrücken, Germany

    Jop F. Sibeyn

Authors
  1. Jop F. Sibeyn
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Igor Prívara Peter Ružička

Rights and permissions

Reprints and permissions

Copyright information

© 1997 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Sibeyn, J.F. (1997). Routing with finite speeds of memory and network. In: Prívara, I., Ružička, P. (eds) Mathematical Foundations of Computer Science 1997. MFCS 1997. Lecture Notes in Computer Science, vol 1295. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0029992

Download citation

  • DOI: https://doi.org/10.1007/BFb0029992

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-63437-9

  • Online ISBN: 978-3-540-69547-9

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics

Search

Navigation