- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi
I have some scripts for runing DART(a meteorology software) with intel compiler and intel mpi library. These scripts were used on cray and MPICH. The environment variable for MPICH setting are :
26 setenv MPICH_VERSION_DISPLAY 1
27 setenv MPICH_ENV_DISPLAY 1
28 setenv MPICH_MPIIO_HINTS_DISPLAY 1
29
30 setenv MPICH_GNI_RDMA_THRESHOLD 2048
31 setenv MPICH_GNI_DYNAMIC_CONN disabled
32
33 setenv MPICH_CPUMASK_DISPLAY 1
34 setenv OMP_NUM_THREADS 1
On original machine 4000 CPU core were used.
On new linux machine with intel MPI, I commet out those variable. When I use 200 CPU core, it takes about 6000 seconds. But when I use 720 CPU core, it takes about 15000 seconds. If I use those environment variable for intel MPI, will the performance improved? How to modify those environment variable? Is there some introduction for setting both intel MPI and MPICH?
Thanks
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
If you have built with OpenMP enabled, you probably don't want to comment out OMP_NUM_THREADS, unless you intend to run 1 MPI rank per node (possibly with HT disabled).

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page