comparison of psmp and popt (with and without openmp)

Samuel Andermatt samuel.a... at student.ethz.ch
Mon Apr 18 07:55:22 UTC 2016

Previous message (by thread): from cp2k 2.4 to 2.5 (stampede) "raman/IR tutorial"
Next message (by thread): Maximum Overlap Method and Geometry Optimization
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

The omp/mpi scaling of CP2K is regularily tested under 
https://dashboard.cp2k.org/archive/scaling/index.html .
I guess the reason for the large speedup through openmp in your case could 
likely be the limited memory per core that you mentioned.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160418/dc1d94c6/attachment.htm>

Previous message (by thread): from cp2k 2.4 to 2.5 (stampede) "raman/IR tutorial"
Next message (by thread): Maximum Overlap Method and Geometry Optimization
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the CP2K-user mailing list