Skip to main content

I/O performance in hybrid MIMD+SIMD machines

  • 3. Computer Science
  • Conference paper
  • First Online:
High-Performance Computing and Networking (HPCN-Europe 1998)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 1401))

Included in the following conference series:

  • 252 Accesses

Abstract

The impact of I/O in the parallel machine is becoming ever more important as the computational power of the processor and the throughput of the communication network are increasing. In this paper we show some scenarios of I/O in parallel machines, in particular we focus on a hybrid MIMD+SIMD architecture which is under the study of an initiative which involves two major MPP vendors in Europe, Alenia Spazio and Meiko. By means of queueing network models we show also how to improve the performance of a SIMD application that alternates I/O and computations in a cyclic fashion.

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

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. I. Foster. Design and Building Parallel Programs. Mathematics and Computer Science Division Argonne National Laboratory. Available at www.mcs.anl.gov/dbpp/text/, 1995.

    Google Scholar 

  2. INTEL CORPORATION, Paragon System User's Guide. Intel SSD, Beaverton, OR, 1995.

    Google Scholar 

  3. S. M. King. Installing, Managing, and Using the IBM AIX Parallel I/O File System. Information Development, IBM Kingston, New York, 1994.

    Google Scholar 

  4. S. J. Lo Verso, M. Isman, A. Nanopoulos, W. Nesheim, E. D. Milne and R. Wheeler, sfs: A parallel file system for the CM-5. In Proceedings of the 1993 Summer USENIX Conference (1993), pages 291–305.

    Google Scholar 

  5. A. Mathis. The PQE1 project, Available at www.enea.it/~hpcn/moshpce/brochure/trends02. html, 1996.

    Google Scholar 

  6. E. L. Miller and R. H. Katz. Input/Output behavior of supercomputing applications. In Proceedings of Supercomputing 91, pages 567–576, November 1991.

    Google Scholar 

  7. B. K. Pasquale and G. Plyzos. A static analysis of I/O characterization of scientific applications in a production workload. In Proceedings of Supercomputing 93, pages 388–397, November 1993.

    Google Scholar 

  8. J.T Poole. Scalable I/O Initiative. California Institute of Technology, Available at www.ccsf.caltech.edu/SIO/, 1996.

    Google Scholar 

  9. E. Smirni, R. A. Aydt, A. A. Chien, and D. A. Reed. I/O requirements of scientific applications: an evolutionary view”. In Fifth IEEE International Symposium on High Performance Distributed Computing, pages 49–59, 1996.

    Google Scholar 

  10. E. Smirni and D. A. Reed. “Workload characterization of input/output intensive parallel applications. PERFROMANCE TOOLS'97 — 9th International Conference on Modelling Techniques and Tools for Computer Performance Evaluation, June 1997.

    Google Scholar 

  11. H. Zima, H.-J. Bast, and M. Gerndt. SUPERB: A tool for semi-automatic MIMD/SIMD parallelization. Parallel Computing, pages 1–18, 1988.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Peter Sloot Marian Bubak Bob Hertzberger

Rights and permissions

Reprints and permissions

Copyright information

© 1998 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Cremonesi, P., Gennaro, C., Marega, R. (1998). I/O performance in hybrid MIMD+SIMD machines. In: Sloot, P., Bubak, M., Hertzberger, B. (eds) High-Performance Computing and Networking. HPCN-Europe 1998. Lecture Notes in Computer Science, vol 1401. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0037196

Download citation

  • DOI: https://doi.org/10.1007/BFb0037196

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-64443-9

  • Online ISBN: 978-3-540-69783-1

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics