A New Hybrid Directory Scheme for Shared Memory Multi-processors

  • Guoteng Pan
  • Lunguo Xie
  • Qiang Dou
  • Erhua He
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3967)


It is increasingly popular to adopt DSM systems to maximize parallelism beyond the limits of SMP. Wherein, a proper cache coherence scheme should be efficient at the memory overhead for maintaining the directories because of its significant impact on overall performance of the system. In this paper, we propose a new hybrid directory scheme which reduces the memory overhead of the directory and improves the memory access time by using our new hybrid directory scheme. We evaluate the performance of our proposed scheme by running six applications on an execution-driven simulator (RSIM). The simulation results show that the performance of a system with hybrid directory can achieve close to that of a multiprocessor with bit-vector directory.


Computer Architecture Cache Coherence Directory Entry Memory Overhead Cache Coherence Protocol 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Eggers, S.J., Katz, R.H.: Evaluating the Performance of Four Snooping Cache Coherency Protocols. In: Proc. of the 16th Annual International Symposium on Computer Architecture (1989)Google Scholar
  2. 2.
    Lenoski, D., Laudon, J.: The directory-based cache coherence protocol for the DASH multiprocessor. In: Proc. of the 17th annual international symposium on Computer Architecture (1990)Google Scholar
  3. 3.
    Censier, L.M., Feautrier, P.: A New Solution to Coherence Problems in Multi-cache Systems. IEEE Transactions on Computers C-27(12), 1112–1118 (1978)CrossRefMATHGoogle Scholar
  4. 4.
    Laudon, J., Lenoski, D.: The SGI Origin: A ccNUMA Highly Scalable Server. In: Proceedings of the 24th international Symposium on Computer Architecture (1997)Google Scholar
  5. 5.
  6. 6.
    Agarwal, A., Simoni, R.: An Evaluation of Directory Schemes for Cache Coherence. In: Proc. of the 15th Znt. Sym. On Computer Architecture, pp. 280–289 (1988)Google Scholar
  7. 7.
    Acacio, et al.: A New Scalable Directory Architecture for Large-Scale Multiprocessors. In: Proceedings of the 7th International Symposium on High-Performance Computer Architecture, pp. 97–106 (2001)Google Scholar
  8. 8.
    Chang, Bhuyan: An Efficient Hybrid Cache Coherence Protocol for Shared Memory Multiprocessors. IEEE Transactions on Computers 48(3), 352–360 (1999)CrossRefGoogle Scholar
  9. 9.
    Deng, R., Xie, L.: Sharing Pattern of Shared Data in Multiprocessors. Computer Engineering and Science 20(A1), 66–69 (1998)Google Scholar
  10. 10.
    Singh, J.P., et al.: SPLASH: Stanford Parallel Applications for Shared-Memory. Computer Architecture News 20, 5–44 (1992)CrossRefGoogle Scholar
  11. 11.
    Woo, S.C., et al.: SPLASH-2 Programs: Characterization and Methodological Considerations. In: Proc. of the 22nd Int’l Symposium on Computer Architecture, pp. 24–36 (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Guoteng Pan
    • 1
  • Lunguo Xie
    • 1
  • Qiang Dou
    • 1
  • Erhua He
    • 1
  1. 1.School of Computer ScienceNational University of Defense TechnologyChangshaP.R. China

Personalised recommendations