I am getting lots of computation errors
log in

Advanced search

Message boards : Number crunching : I am getting lots of computation errors

Author Message
No.15
Send message
Joined: 2 Feb 16
Posts: 13
Credit: 64,229,764
RAC: 0
United States
Message 1164 - Posted: 26 Oct 2017, 13:15:31 UTC

Like 860 of them so far
http://gene.disi.unitn.it/test/result.php?resultid=22183628

Profile valterc
Project administrator
Project tester
Send message
Joined: 30 Oct 13
Posts: 623
Credit: 34,677,535
RAC: 9
Italy
Message 1165 - Posted: 26 Oct 2017, 14:33:57 UTC - in response to Message 1164.

Like 860 of them so far
http://gene.disi.unitn.it/test/result.php?resultid=22183628

Yes, I know, it was a bad batch (with a too low disk space limit for the output file), now I should have correct it, new replicas and new workunits should be okay. I really apologize for this.

Erkan Yilmaz
Send message
Joined: 24 Dec 16
Posts: 17
Credit: 2,367,140
RAC: 0
Germany
Message 1166 - Posted: 26 Oct 2017, 19:24:21 UTC - in response to Message 1165.

Hi Valter,

about 5h passed since your post (and about >3 days since the prob seems to have appeared)...

how is the situation now?

I see there are not enough WUs for users
* the "in progress" are since Oct 24 about 1/3rd of what it used to be :-(


https://steemit.com/boinc/@erkan/no-more-work-at-boinc-project-tn-grid-wus-started-to-decline-about-27h-ago

Erkan

Profile valterc
Project administrator
Project tester
Send message
Joined: 30 Oct 13
Posts: 623
Credit: 34,677,535
RAC: 9
Italy
Message 1169 - Posted: 27 Oct 2017, 9:45:47 UTC - in response to Message 1166.

The 'computation error' problems should slowly be resolved (when all the bad workunits that were distributed error out).

The 'queue' problem is another issue (see http://gene.disi.unitn.it/test/forum_thread.php?id=195&postid=1146). Simply speaking the work generator is not able to satisfy all the users request (we are working on rewriting/optimizing it but it will take some time)

The speed of the work generator depends on the organism dataset (the number of genes, ~3000 Ec, ~28000 Vv), the time needed to complete a workunit depends on the 'structure' of the dataset (Ec is much more 'complicated' than Vv).

Erkan Yilmaz
Send message
Joined: 24 Dec 16
Posts: 17
Credit: 2,367,140
RAC: 0
Germany
Message 1170 - Posted: 27 Oct 2017, 15:21:17 UTC - in response to Message 1169.

Thanks for the quick reply Valter :-)


Post to thread

Message boards : Number crunching : I am getting lots of computation errors


Main page · Your account · Message boards


Copyright © 2024 CNR-TN & UniTN