[Pw_forum] band parallelisation what am I missing?

Tiana Davide davide.tiana at epfl.ch
Mon Sep 21 20:15:08 CEST 2015


Dear all,
I would like to exploit/ test the band parallelisation
I am running a gamma point calculation with nbands 272

I run 3 tests using 272 nband 1, 544 nband 2, 816 nband 3 respectively
Test1
dtiana at daint104:/scratch/daint/dtiana> grep time afm-1.out
     total cpu time spent up to now is       11.8 secs
     total cpu time spent up to now is       22.0 secs
     total cpu time spent up to now is       28.0 secs
Application 5747240 resources: utime ~8080s, stime ~103s, Rss ~173424, inblocks ~11263176, outblocks ~13955496
Test2
dtiana at daint104:/scratch/daint/dtiana> grep time afm-2.out
     total cpu time spent up to now is       15.1 secs
     total cpu time spent up to now is       27.3 secs
     total cpu time spent up to now is       33.9 secs
Application 5747236 resources: utime ~17984s, stime ~209s, Rss ~175072, inblocks ~22506882, outblocks ~27724272
Test3
dtiana at daint104:/scratch/daint/dtiana> grep time afm-3.out
     total cpu time spent up to now is       20.5 secs
     total cpu time spent up to now is       30.4 secs
     total cpu time spent up to now is       36.5 secs
Application 5747233 resources: utime ~26335s, stime ~309s, Rss ~176636, inblocks ~33763645, outblocks ~41493055

I am sure I am doing something stupid but I cannot understand what.
Can anyone give me a tip?

Cheers
Davide

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150921/9ac3b281/attachment.html>


More information about the users mailing list