Parallel Performance of MPI Sorting Algorithms on Dual - Core Processor Windows-Based Systems

Free registration required

Executive Summary

Message Passing Interface (MPI) is the most commonly used paradigm in writing parallel programs since it can be employed not only within a single processing node but also across several connected ones. MPI standard has been designed to enhance portability in parallel applications, as well as to bridge the gap between the performance offered by a parallel architecture and the actual performance delivered to the application. Two critical areas determine the overall performance level of an MPI implementation. The first area is the low-level communication layer that the upper layers of an MPI implementation can use as foundations. The second area covers the communication progress and management.

  • Format: PDF
  • Size: 444 KB