next up previous
Next: Results for the Up: Portability versus Efficiency?

Previous: Algorithm Architecture on

Performance Comparison

We run two application programs (15-puzzle and floorplan optimization) on three machines (GCel, GC/PP, UNIX workstation cluster) and two programming models (Parix and PVM). Not counting the workstation experiments, for which only PVM was available, we run a total of 10 benchmarks. From the bulk of statistical data, we selected the most significant results to be presented in the following sections.

Tue May 16 19:29:30 MET DST 1995