MPIMMU — Providing a Single Address Space over a Cluster


Distributed shared memory combines the shared memory programming model with physically distributed memory machines. In this paper we present our approach in implementing a software distributed shared memory system that runs on a cluster and provides access to a single address space. It's architecture is based on a modular design where every module performs a specific task: communication over the network, caching or local memory operations. Pthreads and MPI implementation details are provided along with performance results for local and remote memory operations with and without caching.

  • Abstract
  • Key Words
  • 1 Introduction
  • 2 Related Work
  • 3 Implementation
  • 4 Test Cases and Results
  • 5 Conclusions and Future Work
  • Acknowledgments
  • References

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In