Advertisement

Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

Out-of-order execution in sequentially consistent shared-memory systems: Theory and experiments

Abstract

Traditional implementation of sequential consistency in shared-memory systems requires memory accesses to be globally performed in program order. Based on an event ordering model for correct executions in shared-memory systems, this paper proposes and proves that out-of-order execution does not influence the correctness of an execution providing certain condition is met. Simulation results show that out-of-order execution proposed in this paper is an effective way to improve the performance of a sequentially consistent shared-memory system.

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

References

  1. [1]

    Lamport L. How to make a multiprocessor computer that correctly executes multiprocessor programs.IEEE Transactions on Computers, 1979, C-28(9): 690–691.

  2. [2]

    Gharachorloo K, Lenoski D, Laudon Jet al. Memory consistency and event ordering in scalable shared-memory multiprocessors. InProceedings of the 17th Annual International Symposium on Computer Architecture, 1990, pp.15–26.

  3. [3]

    Shasha D, Snir M. Efficient and correct execution of parallel programs that share memory.ACM Transactions on Programming Languages and System, 1988, 10(2): 282–312.

  4. [4]

    Scheurich C, Dubois M. Correct memory operation of cached-based multiprocessors. InProceedings of the 14th Annual International Symposium on Computer Architecture, 1987, pp.234–243.

  5. [5]

    Coller W. Reasoning About Parallel Architectures. Englewood Cliffs (ed.), NJ: Prentice-Hall, 1992.

  6. [6]

    Archibald J, Baer J. Cache coherence protocols: evaluation using a multiprocessor simulation model.ACM Transactions on Computer Systems, 1986, 4: 273–298.

  7. [7]

    Tomasevic M, Milutinovic V. A simulation study of snoopy cache coherence protocols. InProceedings of the 25th Hawaii International Conference on System Sciences, 1992, pp.427–436.

  8. [8]

    Dubios M, Briggs F. Effects of csche coherence in multiprocessors.IEEE Transaction on Computers, 1982, C-31(11): 1083–1099.

  9. [9]

    Dubios M, Scheurich C, Briggs F. Memory access buffering in multiprocessors. InProceedings of the 13th International Symposium on Computer Architecture, 1986, pp.434–442.

  10. [10]

    Dubios M, Scheurich C, Briggs F. Synchronization, coherence, and event ordering in multiprocessors.Computer, 1988, 21(2): 9–21.

  11. [11]

    Adve S, Hill M. Weak Ordering: A new definition. InProceedings of the 17th Annual International Symposium on Computer Architecture, 1990, pp.1–14.

  12. [12]

    Goodman J. Cache Consistency and sequential consistency.Technical Report No. 61, SCI Committee, March 1989.

  13. [13]

    Gharachorloo K, Gupta A, Hennessy J. Two techniques to enhance the performance of memory consistency models. InProceedings of the 1991 International Conference on Parallel Processing, 1991, pp.I-335-I-364.

  14. [14]

    Herbert B. Enderton. Elements of Set Theory. London: Academic Press, Inc. Ltd., 1977.

  15. [15]

    Lenoski D, Laudon J, Gharachorloo Get al. The directory-based cache coherence protocol for the DASH multiprocessors. InProceedings of the 17th Annual International Symposium on Computer Architecture, 1990, pp.148–158.

Download references

Author information

Correspondence to Weiwu Hu.

Additional information

The work is supported by the President Young Investigator Foundation of Chinese Academy of Sciences and the National Natural Science Foundation of China.

For the biography ofHu Weiwu please refer to p.109, No.2, Vol.13 of this Journal.

Xia Peisu in an Academician of the Chinese Academy of Sciences and a Professor of the Institute of Computing Technology. She received her Ph.D. degree in electrical engineering from University of Edinburgh, United Kindom, in 1950. Her current research interests include computer architecture, computer engineering, high speed pipeline system and parallel processing.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hu, W., Xia, P. Out-of-order execution in sequentially consistent shared-memory systems: Theory and experiments. J. of Comput. Sci. & Technol. 13, 125–140 (1998). https://doi.org/10.1007/BF02946601

Download citation

Keywords

  • Shared memory
  • sequential consistency
  • event ordering
  • write atomic, out-of-order execution, simulation