Page 1 of 1

Project: 2669 (Run 8, Clone 87, Gen 141)

Posted: Sun Oct 11, 2009 10:45 am
by Zagen30
This WU has an unusual number of steps and runs very slowly:

Code: Select all

[16:45:32] + Processing work unit
[16:45:32] At least 4 processors must be requested.Core required: FahCore_a2.exe
[16:45:32] Core found.
[16:45:33] Working on queue slot 06 [October 11 16:45:33 UTC]
[16:45:33] + Working ...
[16:45:33] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a2.exe -dir work/ -suffix 06 -checkpoint 15 -verbose -lifeline 5047 -version 624'

[16:45:33] 
[16:45:33] *------------------------------*
[16:45:33] Folding@Home Gromacs SMP Core
[16:45:33] Version 2.10 (Sun Aug 30 03:43:28 CEST 2009)
[16:45:33] 
[16:45:33] Preparing to commence simulation
[16:45:33] - Ensuring status. Please wait.
[16:45:34] Called DecompressByteArray: compressed_data_size=4829642 data_size=23976217, decompressed_data_size=23976217 diff=0
[16:45:36] - Digital signature verified
[16:45:36] 
[16:45:36] Project: 2669 (Run 8, Clone 87, Gen 141)
[16:45:36] 
[16:45:36] Assembly optimizations on if available.
[16:45:36] Entering M.D.
[16:45:48] Run 8, Clone 87, Gen 141)
[16:45:48] 
[16:45:48] Entering M.D.
NNODES=4, MYRANK=1, HOSTNAME=kyle-folding
NNODES=4, MYRANK=0, HOSTNAME=kyle-folding
NODEID=0 argc=20
NNODES=4, MYRANK=2, HOSTNAME=kyle-folding
NODEID=2 argc=20
NODEID=1 argc=20
Reading file work/wudata_06.tpr, VERSION 3.3.99_development_20070618 (single precision)
NNODES=4, MYRANK=3, HOSTNAME=kyle-folding
NODEID=3 argc=20
Note: tpx file_version 48, software version 68

NOTE: The tpr file used for this simulation is in an old format, for less memory usage and possibly more performance create a new tpr file with an up to date version of grompp

Making 1D domain decomposition 1 x 1 x 4
starting mdrun '22860 system'
35500004 steps,  71000.0 ps (continuing from step 35002468,  70004.9 ps).
[17:27:33] pleted 4976 out of 497536 steps  (1%)
[18:10:09] Completed 9951 out of 497536 steps  (2%)
[18:51:59] Completed 14927 out of 497536 steps  (3%)
[19:16:31] - Autosending finished units... [October 11 19:16:31 UTC]
[19:16:31] Trying to send all finished work units
[19:16:31] + No unsent completed units remaining.
[19:16:31] - Autosend completed
[19:34:26] Completed 19902 out of 497536 steps  (4%)
[20:16:23] Completed 24877 out of 497536 steps  (5%)
[20:58:32] Completed 29853 out of 497536 steps  (6%)
[21:40:30] Completed 34828 out of 497536 steps  (7%)
[22:21:37] Completed 39803 out of 497536 steps  (8%)
[23:03:17] Completed 44779 out of 497536 steps  (9%)
[23:44:39] Completed 49754 out of 497536 steps  (10%)
[00:26:49] Completed 54729 out of 497536 steps  (11%)
[01:07:55] Completed 59705 out of 497536 steps  (12%)
[01:16:31] - Autosending finished units... [October 12 01:16:31 UTC]
[01:16:31] Trying to send all finished work units
[01:16:31] + No unsent completed units remaining.
[01:16:31] - Autosend completed
[01:48:18] Completed 64680 out of 497536 steps  (13%)
[02:30:32] Completed 69656 out of 497536 steps  (14%)
[03:11:53] Completed 74631 out of 497536 steps  (15%)
[03:53:23] Completed 79606 out of 497536 steps  (16%)
[04:36:46] Completed 84582 out of 497536 steps  (17%)
[05:17:27] Completed 89557 out of 497536 steps  (18%)
[05:58:59] Completed 94532 out of 497536 steps  (19%)
[06:40:48] Completed 99508 out of 497536 steps  (20%)
To save people the math, that's around 40 minutes per frame, when this computer (2.0 GHz C2D) normally averages around 20. This is a virtual Linux Box running Ubuntu 9.04 with kernel 2.6.28.7.

I didn't find any reports about this one, but if this belongs in another thread go ahead and merge.