Using Speculative Push for Unnecessary Checkpoint Creation Avoidance

  • Arkadiusz Danilecki
  • Michał Szychowiak
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4025)


This paper discusses a way of incorporating speculation techniques into Distributed Shared Memory (DSM) systems with checkpointing mechanism without creating unnecessary checkpoints. Speculation is a general technique involving prediction of the future of a computation, namely accesses to shared objects unavailable on the accessing node (read faults). Thanks to such predictions objects can be pushed to requesting nodes before the actual access operation is performed, resulting, at least potentially, in a considerable performance improvement. This mechanism is a foundation for the proposed SpecCkpt protocol based on independent checkpointing integrated with a coherence protocol for a given consistency model introducing little overhead. It ensures the consistency of checkpoints, at the same time allowing a fast recovery from failures.


Shared Object Distribute Share Memory Read Access Speculation Technique Distribute Share Memory System 
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.


  1. 1.
    Bianchini, R., Pinto, R., Amorim, C.L.: Data Prefetching for Software DSMs. In: Proc. Int. Conference on Supercomputing, Melbourne, Australia (1998)Google Scholar
  2. 2.
    Carbajal, J., Michael, M., Nguyen, A.-T., Torrellas, J., Sharma, A.: Augmint: A Multiprocessor Simulation Environment for Intel x86 Architectures. CSRD Technical Report 1463 (March 1996)Google Scholar
  3. 3.
    Danilecki, A., Szychowiak, M.: Checkpointing Speculative Distributed Shared Memory. In: Wyrzykowski, R., Dongarra, J., Meyer, N., Waśniewski, J. (eds.) PPAM 2005. LNCS, vol. 3911, pp. 9–16. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  4. 4.
    Danilecki, A., Szychowiak, M.: Checkpointing Speculative DSM Systems, Research Report RA-021/05, Institute of Computing Science, Poznan University of Technology (2005)Google Scholar
  5. 5.
    Danilecki, A., Szychowiak, M., Kobusinski, J.: Simplified DSM simulation with the use of the Augmint backend, Research Report RA-04/06, Institute of Computing Science, Poznan University of Technology (2006)Google Scholar
  6. 6.
    Gupta, A., Ohara, M., Singh, J., Torrie, E., Woo, S.: The SPLASH2 Programs: Characterization and Methodological Considerations. In: Proc. 22nd Int. Symposium on Computer Architecture (ISCA 1995) (May 1995)Google Scholar
  7. 7.
    Kongmunvattana, A., Tanchatchawal, S., Tzeng, N.-F.: Coherence-Based Coordinated Checkpointing for Software Distributed Shared Memory Systems. In: Proc. 20th Conference on Distributed Computing Systems, pp. 556–563 (2000)Google Scholar
  8. 8.
    Lai, A.-C., Babak Falsafi, B.: Selective, Accurate, and Timely Self-Invalidation Using Last-Touch Prediction. In: Proc. 27th Int. Symposium on Computer Architecture (ISCA 27), Vancouver, BC, Canada, pp. 139–148 (2000)Google Scholar
  9. 9.
    Park, T., Yeom, H.: A Low Overhead Logging Scheme for Fast Recovery in Distributed Shared Memory Systems. Journal of Supercomputing 15(3), 295–320 (2002)CrossRefMATHGoogle Scholar
  10. 10.
    Rajwar, R., Kagi, A., Goodman, J.R.: Inferential Queueing and Speculative Push. International Journal of Parallel Programming (IJPP) 32(3), 273–284 (2004)MATHGoogle Scholar

Copyright information

© IFIP International Federation for Information Processing 2006

Authors and Affiliations

  • Arkadiusz Danilecki
    • 1
  • Michał Szychowiak
    • 1
  1. 1.Institute of Computing SciencePoznań University of TechnologyPoznańPoland

Personalised recommendations