Skip to main content
SpringerLink
Log in

Experiences with the Mether-NFS virtual shared memory system

  • Parallal Programming Enviroments
  • Conference paper
  • First Online:
High-Performance Computing and Networking (HPCN-Europe 1994)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 797))

Included in the following conference series:

  • 152 Accesses

Abstract

Workstation clusters have recently attracted high interest as a technology providing supercomputer class performance at much lower price levels. The message passing programming model dominates the application development, despite the overhead and the complexity introduced by the explicitly coded synchronisation and data transfers. We give an introduction to the virtual shared memory programming model and report on the experiences with an implementation running on an FDDI network with 8 machines. We show that the direct coherence control mechanisms provided by the Mether-NFS system can be used to overcome the false sharing overhead caused by the coherency protocol. They provide an excellent tool for parallelizing programs using complex data structures with dynamic load balancing.

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. Bolosky, William J., Scott, Michael L.: False Sharing and its Effect on Shared Memory Performance. In: Proc. Distributed & Multiprocessor Systems (SEDMS IV), USENIX Association, Berkeley, CA 94710, September (1993) 57–71.

    Google Scholar 

  2. Brooks, Eugene D.: Massively Parallel Computing. In: The 1992 MPCI Yearly Report: Harnessing the Killer Micros. Lawrence Livermore National Laboratory, Livermore, CA 94550, August (1992) 1–6.

    Google Scholar 

  3. Cremer, M.: O-D Estimation: Dynamic Methods. In: M. Papageorgiou, editor, Concise Encyclopedia of Traffic and Transportation Systems. Pergamon Press, Oxford, (1990).

    Google Scholar 

  4. Daganzo, Carlos F.: The Cell transmission model: A dynamic representation of Highway traffic consistent with the hydrodynamic theory. Technical report, Department of Civil Engeneering and Institute of Transportation, University of California, Berkeley CA 94720, July (1993).

    Google Scholar 

  5. David, A., Rosenthal, S. H.: Evolving the Vnode interface. In: Usenix Conference Proceedings, Anaheim, CA, Summer (1990) 107–117.

    Google Scholar 

  6. Dongarra, Jack, Geist, G. A., Mancheck, Robert, Sunderam, V. S.: Integrated PVM Framework supports Heterogeneous Network Computing. Computers in Physics, 7(2) Mar/Apr (1993).

    Google Scholar 

  7. Heddaya, Abdesalam, Sinha, Himanshu: Coherence, Non-coherence and Local Consistency in Distributed Shared memory for Parallel Computing. Technical Report BU-CS-92-004, Computer Science Department, Boston University, Boston, MA 02215, USA, May (1992).

    Google Scholar 

  8. Keller, Hartmut, Ploss, Gerhard: Real-Time Identification of O-D Network Flows from Counts for Urban Traffic Control. In: N. H. Gartner and N. H. M. Wilson, editors, Transportation and Traffic Theory. Elsevier Science Publishing Co., (1987) 267–285.

    Google Scholar 

  9. Kentemich, Thomas, Solchenbach, Karl: Paralleles Rechnen auf Workstations. KFA-ZAM-IB-9224, Forschungszentrum Jülich ZAM, Germany, November (1992).

    Google Scholar 

  10. Kleinman, Steven R.: Vnodes: An Architecture for Multiple File System Types in Sun UNIX. In: Usenix Conference Proceedings, Atlanta, GA, Summer (1986) 238–247

    Google Scholar 

  11. Li, Kai: Shared Virtual Memory on Loosely Coupled Multiprocessors. Ph.D. thesis, Yale University, Department of Computer Science, September (1986)

    Google Scholar 

  12. Nagel, K., Schleicher, A.: Microscopic traffic modeling on parallel high performance computers. Parallel Computing. 20 (1994) 125–146

    Google Scholar 

  13. Nagel, K., Schreckenberg, M.: A cellular automaton model for freeway traffic. J. Physic I France. 2 (1992) 2221–2229.

    Google Scholar 

  14. Minnich, Ronald G.: Mether-NFS: A modified NFS Which Supports Virtual Shared Memory. In Proc. Distributed & Multiprocessor Systems (SEDMS IV), USENIX Association, Berkeley, CA 94710, September (1993) 89–107.

    Google Scholar 

  15. Oed, Wilfried: The Cray Research Massively Parallel Processor System CRAY T3D. Technical report, Cray Research GmBH, 80922 München, Germany, November (1993).

    Google Scholar 

  16. Ploss, Gerhard et.al.: Systemdynamische Schätzung der Matrix der Verkehrsbeziehungen in Außerortsstraßen als Grundlage für die Steuerung von Verkehrsleit-systemen. Forschungsauftrag 1.124 R89E, Forschungsgesellschaft für Straßen-und Verkehrswesen, 50996 Köln, Germany, May (1992).

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Zentrum für Paralleles Rechnen, Universität zu Köln, Weyertal 80, 50931, Köln, Germany

    Jörg-Thomas Pfenning

Authors
  1. Jörg-Thomas Pfenning
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Wolfgang Gentzsch Uwe Harms

Rights and permissions

Reprints and permissions

Copyright information

© 1994 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Pfenning, JT. (1994). Experiences with the Mether-NFS virtual shared memory system. In: Gentzsch, W., Harms, U. (eds) High-Performance Computing and Networking. HPCN-Europe 1994. Lecture Notes in Computer Science, vol 797. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-57981-8_136

Download citation

  • DOI: https://doi.org/10.1007/3-540-57981-8_136

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-57981-6

  • Online ISBN: 978-3-540-48408-0

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics

Search

Navigation