Problem with finishing of cp2k jobs
Shadrina Maria
shadr... at gmail.com
Thu Jul 28 19:19:03 UTC 2011
Dear All,
I have a problem with finishing of cp2k jobs. Jobs have frozen after the
last section relevant to calculation and before the section "MULTIGRID INFO"
at the end of the output file:
-------------------------------------------------------------------------------
---- MULTIGRID INFO
----
-------------------------------------------------------------------------------
count for grid 1: 186800 cutoff [a.u.]
150.00
count for grid 2: 24992 cutoff [a.u.]
37.50
count for grid 3: 9908 cutoff [a.u.]
9.38
count for grid 4: 952 cutoff [a.u.]
2.34
total gridlevel count : 222652
-------------------------------------------------------------------------------
- MESSAGE PASSING PERFORMANCE
-
-------------------------------------------------------------------------------
ROUTINE CALLS TOT TIME [s] AVE VOLUME [Bytes] PERFORMANCE
[MB/s]
MP_Group 6 0.000
MP_Bcast 3696 0.133 371057.
10342.59
MP_Allreduce 6465 11.179 38542.
22.29
.......
......
......
I found that the last subroutine called in these jobs is
pw_pool_give_back_pw. The input file is also attached.
It seems that freezing of jobs depends on number of cores somehow. Jobs on
1x8 have completed correctly and jobs on 1x4 and 2x8 have frozen.
Have you any suggestions what the problem could be?
For any case:
CP2K version 2.2.262
the lib is MKL-Scalapack
the system is cluster of XEON nodes (8cores/node) with Infiniband switch
the compiler is Intel's
Thank you in advance.
Best regards, Maria.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20110728/21769ceb/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.inp
Type: application/octet-stream
Size: 2556 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20110728/21769ceb/attachment.obj>
More information about the CP2K-user
mailing list