View Single Post
Old 2020-08-15, 15:58   #22
VBCurtis's Avatar
Feb 2005
Riverside, CA

113916 Posts

Originally Posted by frmky View Post
Here's a bench using compute nodes with one Xeon E5-2650 v4 Broadwell cpu with 12-cores, 24 threads.

1 node 7h 40m
2 nodes 2h 45m
4 nodes 1h 35m
8 nodes 1h 10m

Not sure why the time for one node is so high compared to the others? Perhaps something fitting into the cache with the smaller matrices on each node?
It never occured to me that MPI - 2 nodes would be more than twice as fast, under any test. Neat! Now, if only ubuntu would fix MPI....
VBCurtis is online now   Reply With Quote