Skip to main content

Fair metacomputer — FMC

  • Conference paper
  • First Online:
  • 143 Accesses

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 879))

Abstract

Networks of powerful workstations are much more popular than large mainframes, or even supercomputers. But it is very hard to gain good utilisation of network. The purpose of connecting workstations together as a Fair Metacomputer (FMC) is to give unused power of application-users to power users. Metacomputer by definition takes care of distributing parts of parallel programs over the network and communication among them. FMC makes a step forward. It tries to be as fair as possible to all users. It uses only idle workstations for tasks execution. When an idle workstation becomes busy (user returned to work), tasks migrate to other idle workstations.

FMC can be used in any computer network, where intensive calculations take place. It is especially useful in areas, where problems can be paralleled — divided in few parts, that run at the same time (chemistry, physics, mathematics, ...).

Another function of FMC is important. When a workstation is suddenly shotdown, or crashed, programs are not lost. At start-up, programs will run from where they were interrupted. Only a small part of computation is lost.

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

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. CONVEX Meta Series System and Network Management: Concepts, CONVEX Press, Richardson, Texas, 1992.

    Google Scholar 

  2. A. Bricker, M. Litzkow, M. Livny, Condor Technical Summary, Computer Science Department, University of Wisconsin — Madison, 1992.

    Google Scholar 

  3. S.A.Rago, UNIX System V Network Programming, Addison-Wesley Publishing, 1993.

    Google Scholar 

  4. M.Litzkow, M.Solomon, Supporting Checkpointing and Process Migration Outside the UNIX Kernel, Poreceedings of Usenix Winter Conference, San Francisco, California, 1992.

    Google Scholar 

  5. L. Reiss, J. Radin, UNIX System Administration Guide, pg. 221–245, Osborne McGraw-Hill, New York, 1993.

    Google Scholar 

  6. F.Douglis, J.K.Ousterhout, M.F.Kaashoek, A.S.Tanenbaum, Comperison of Two Distributed Systems: Amoeba nad Sprite, Computing Systems, 1991.

    Google Scholar 

  7. S.Son, B.Nelson, J.Archibald, Efficient Utilization of Distributed Workstation Resources, Department of Electrical and Computer Engineering, Brigham Young University, 1991.

    Google Scholar 

  8. A.Beguelin, J.Dongarra, A.Geist, R.Manchek, V.Sunderman, A User's Guide to PVM — Parallel Virtual Machine.

    Google Scholar 

  9. A.Bjornson, Parallel Algorithm and Application Development Techniques, Proceedings of the Scalable Computing Working Group, Munchen, 1993.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Jack Dongarra Jerzy Waśniewski

Rights and permissions

Reprints and permissions

Copyright information

© 1994 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Zrimsek, A., Smukavec, T. (1994). Fair metacomputer — FMC. In: Dongarra, J., Waśniewski, J. (eds) Parallel Scientific Computing. PARA 1994. Lecture Notes in Computer Science, vol 879. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0030182

Download citation

  • DOI: https://doi.org/10.1007/BFb0030182

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-58712-5

  • Online ISBN: 978-3-540-49050-0

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics