HPC tutorial heading HPC tutorial heading
Last update: 03/22/2006

HPC tutorial contents

Speedup data from hybrid MPI-OpenMP

4,000 cycles

Note: The current version of this program is set up to run only two MPI tasks. This will be modified in the next version.

No. threads

No. tasks


1 2 2.1
2 2 1.29

This version of the hybrid code runs with two MPI tasks in 2.1 seconds. With one thread added, it runs in 1.29 seconds, a 61% speedup.

If you have questions about this document, please contact SCD Customer Support . You can also reach us by telephone 24 hours a day, seven days a week at 303-497-1278. Additional contact methods: consult1@ucar.edu and during business hours in NCAR Mesa Lab Suite 39.

© Copyright 2006. University Corporation for Atmospheric Research (UCAR). All Rights Reserved.