Source Code and Task Graphs in Program Optimization

  • Authors:
  • Welf Löwe;Wolf Zimmermann;Sven Dickert;Jörn Eisenbiegler

  • Affiliations:
  • -;-;-;-

  • Venue:
  • HPCN Europe 2001 Proceedings of the 9th International Conference on High-Performance Computing and Networking
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present scheduling techniques considering the source code as well as the task graph of a parallel program. This approach allows to chose the proper scheduling algorithm and to estimate the execution time of the target program. We demonstrate this on stencil computations and their optimizations.