Advertisement

Application-Controlled Coherence Protocols for Scope Consistent Software DSMs

  • Cicero Roberto de Oliveira Galdino
  • Alba Cristina Magalhaes Alves de Melo
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2110)

Abstract

To use the shared memory programming paradigm in distributed architectures where there is no physically shared memory, an abstraction must be created. This abstraction is known as Distributed Shared Memory (DSM). To reduce communication costs, DSM systems usually replicate data. This approach generates a coherence problem, which is generally solved by a memory coherence protocol. Unfortunately, it seems that there is no coherence protocol that achieves good performance for a large set of applications since the most appropriate coherence protocol depends on how the application accesses data. For this reason, it is interesting for a DSM system to provide multiple coherence protocols. This article presents and evaluates a low-overhead mechanism that allows a DSM application to choose among multiple coherence protocols. This mechanism was incorporated in JIAJIA, a DSM system that implements scope consistency with a write-invalidate protocol. Our results on some benchmarks show a significant reduction on the number of messages exchanged, leading to better performance results.

Keywords

Access Pattern Home Node Distribute Shared Memory Cache Coherence Protocol Page State 
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. 1.
    Amza C., Cox A., Dwarkakas S., Zwaenenpoel W.: Software DSM Protocols that Adapt between Single Writer and Multiple Writer, Proc. of HPCA’97, p. 261–271, February, 1997.Google Scholar
  2. 2.
    Bailey D., Barton J., Lasinski T., Simon H.: The NAS Parallel Benchmarks, Technical Report 103863, NASA, July, 1993.Google Scholar
  3. 3.
    Blumrich M., Li K., Alpert R., Dubnicki C., Felten E.: A Virtual Memory Mapped Network Interface for the SHRIMP Multicomputer, Proc of ISCA’94, p. 142–153, April, 1994.Google Scholar
  4. 4.
    Carter J., Efficient Distributed Shared Memory Based on Multi-Protocol Release Consistency. PhD Thesis, Rice University, January, 1994.Google Scholar
  5. 5.
    Culler D., Singh J., Gupta A.: Parallel Computer Architecture: A Hardware/Software Approach, Morgan Kaufmann, 1998.Google Scholar
  6. 6.
    Gharachorloo K.: The Plight of Software Distributed Shared Memory, Proc. Of the 1st ACM International Workshop on Software DSM, Rhodes, Greece, June, 1999.Google Scholar
  7. 7.
    Hu W., Shi W., Tang Z.: JIAJIA: An SVM System Based on A New Cache Coherence Protocol. In Proc. of HPCN’99, LNCS 1593, pp. 463–472, Springer-Verlag, April, 1999.Google Scholar
  8. 8.
    Iftode L.: Home-Based Shared Virtual Memory, PhD Thesis, June, 1998, Princeton University, Dept of Computer Science, 132 pages.Google Scholar
  9. 9.
    Iftode L., Singh J., Li K.: Scope Consistency: Bridging the Gap Between Release Consistency and Entry Consistency, Proc. Of the 8th ACM SPAA’96, June, 1996, pages 277–287.Google Scholar
  10. 10.
    Keleher P., Cox A., Dwarkakas S., Zwaenenpoel W.: TreadMarks: Distributed Shared Memory on Standard Workstations and Operating Systems”, Proc. of USENIX, January, 1994, pages 115–132.Google Scholar
  11. 11.
    Keleher P: Tapeworm: High-Level Abstraction of Shared Accesses, Proc of the 3rd Symposium on Operating Systems Design and Implementation, February, 1999.Google Scholar
  12. 12.
    Lee J., Jhon C.: Reducing Coherence Overhead of Barrier Synchronization in Software DSMs, Proc. Of the 1998 ACM Int. Conf. on Supercomputing, November, 1998.Google Scholar
  13. 13.
    Monnerat L., Bianchinni R.: Efficiently Adapting to Sharing Patterns in Software DSMs, Proc. Of the 4th HPCA’98, February, 1998.Google Scholar
  14. 14.
    Mosberger D.: Memory Consistency Models, Operating Systems Review, p. 18–26, 1993.Google Scholar
  15. 15.
    Shi W.: Improving the Performance of DSM Systems, PhD Thesis, Chinese Academy of Sciences, November, 1999.Google Scholar
  16. 16.
    Singh J., Weber W., Gupta A.: SPLASH: Stanford Parallel Applications for Shared Memory, Computer Architecture News, 20(1):5–44, March, 1992.CrossRefGoogle Scholar
  17. 17.
    Speight E., Bennett J.: Reducing Coherence-Related Communication in Software Distributed Shared Memory Systems, Technical Report ECE-TR-98-03, Rice University,1998.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Cicero Roberto de Oliveira Galdino
    • 2
  • Alba Cristina Magalhaes Alves de Melo
    • 1
  1. 1.Department of Computer ScienceUniversity of BrasiliaBrasiliaBrazil
  2. 2.Medical Informatics GroupSarah Hospital NetworkBrasíliaBrazil

Personalised recommendations