Advanced search

Message boards : Number crunching : Saving a perfectly good WU from the oblivion...

Author Message
Bedrich Hajek
Send message
Joined: 28 Mar 09
Posts: 485
Credit: 10,408,398,466
RAC: 14,655,175
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 43968 - Posted: 15 Jul 2016 | 1:49:36 UTC

Saving a perfectly good WU from the oblivion of being labeled a "too many errors may have a bug" WU, this may have been one of them:

name e1s29_1-GERARD_FXCXCL12RX_1153966_2-0-1-RND0349
application Long runs (8-12 hours on fastest card)
created 29 Jun 2016 | 10:20:06 UTC
canonical result 15202031
granted credit 267,900.00
minimum quorum 1
initial replication 1
max # of error/total/success tasks 7, 10, 6
Task
click for details Computer Sent Time reported
or deadline
explain Status Run time
(sec) CPU time
(sec) Credit Application
15178821 196423 29 Jun 2016 | 16:57:27 UTC 1 Jul 2016 | 12:34:19 UTC Aborted by user 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15182848 347193 1 Jul 2016 | 14:14:13 UTC 6 Jul 2016 | 14:14:13 UTC Timed out - no response 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15190855 349929 6 Jul 2016 | 16:35:30 UTC 8 Jul 2016 | 16:51:29 UTC Error while computing 74,715.23 62,929.35 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15194032 350898 8 Jul 2016 | 18:25:07 UTC 8 Jul 2016 | 22:23:38 UTC Error while computing 1,363.09 503.14 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15194309 228333 8 Jul 2016 | 23:53:52 UTC 9 Jul 2016 | 0:13:52 UTC Error while computing 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15194444 351065 9 Jul 2016 | 2:15:50 UTC 14 Jul 2016 | 2:15:50 UTC Timed out - no response 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15202031 263612 14 Jul 2016 | 2:16:21 UTC 14 Jul 2016 | 11:40:18 UTC Completed and validated 26,244.25 26,103.89 267,900.00 Long runs (8-12 hours on fastest card) v8.48 (cuda65)


I was just wondering how many perfectly good WUs have the unfortunate fate of being labeled that, just because they are sent to bad hosts?


And more importantly, how does that effect the project?


Every so often, I do end up successfully completing one of these WUs.




Bedrich Hajek
Send message
Joined: 28 Mar 09
Posts: 485
Credit: 10,408,398,466
RAC: 14,655,175
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 44088 - Posted: 3 Aug 2016 | 0:01:43 UTC

Here is another example:

name e25s8_e16s6p0f166-GERARD_CXCL12VOLKDIM_40310216_2-0-1-RND9615
application Long runs (8-12 hours on fastest card)
created 23 Jul 2016 | 5:05:43 UTC
canonical result 15223697
granted credit 257,100.00
minimum quorum 1
initial replication 1
max # of error/total/success tasks 7, 10, 6
Task
click for details Computer Sent Time reported
or deadline
explain Status Run time
(sec) CPU time
(sec) Credit Application
15213021 243850 23 Jul 2016 | 5:06:56 UTC 23 Jul 2016 | 5:09:12 UTC Error while computing 1.13 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15213039 192505 23 Jul 2016 | 5:09:24 UTC 23 Jul 2016 | 5:13:32 UTC Error while computing 1.11 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15213061 126289 23 Jul 2016 | 5:13:40 UTC 23 Jul 2016 | 5:29:07 UTC Error while computing 1.44 0.00 --- Long runs (8-12 hours on fastest card) v8.41 (cuda60)
15213111 234149 23 Jul 2016 | 5:29:45 UTC 28 Jul 2016 | 5:29:45 UTC Timed out - no response 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15219191 327276 28 Jul 2016 | 5:29:55 UTC 28 Jul 2016 | 5:31:20 UTC Error while computing 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15219220 346632 28 Jul 2016 | 5:31:28 UTC 28 Jul 2016 | 5:38:29 UTC Error while computing 1.11 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15219230 206399 28 Jul 2016 | 5:38:32 UTC 1 Aug 2016 | 1:55:12 UTC Abandoned 0.00 0.00 --- Long runs (8-12 hours on fastest card) v8.48 (cuda65)
15223697 30790 1 Aug 2016 | 1:55:30 UTC 1 Aug 2016 | 13:13:06 UTC Completed and validated 22,555.91 22,210.44 257,100.00 Long runs (8-12 hours on fastest card) v8.48 (cuda65)


I wonder if I'll ever get answers on the questions in the previous post?


Post to thread

Message boards : Number crunching : Saving a perfectly good WU from the oblivion...

//