High Performance Computing in Science and Engineering ‘13 by Wolfgang E. Nagel Dietmar H. Kröner & Michael M. Resch

Author:Wolfgang E. Nagel, Dietmar H. Kröner & Michael M. Resch
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

5.1 Performance of Previous and Current MPI Implementation

We first compare the performance of the current MPI implementation with the initial status of the code before optimization to demonstrate our efforts until now. The reference test case consists of a cubic computational domain with 1283 grid cells and represents a basic fluid dynamic process, namely a droplet in a free-stream. For every computation there was an equal amount of MPI processes in the three spatial directions, where each MPI process was assigned to one block of cubic shape. For this investigation only MPI parallelization is used. The resulting speedups as well as the ideal speedup are depicted in Fig. 10.

Fig. 10Speedup comparison of the initial and the current status of MPI optimizations

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.