Advanced search

Message boards : Number crunching : I652-SANTI_baxbimSPW-0-62-RND0982_0

Author Message
Killersocke
Send message
Joined: 18 Oct 13
Posts: 53
Credit: 406,647,419
RAC: 0
Level
Gln
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34038 - Posted: 27 Nov 2013 | 16:22:03 UTC
Last modified: 27 Nov 2013 | 16:25:15 UTC

I652-SANTI_baxbimSPW-0-62-RND0982_0
Packet =4948744

This file is now over 6 hrs crunching
ready only 52%
Display Resttime: 1:35
This could'nt be

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,600,211,851
RAC: 8,778,017
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34039 - Posted: 27 Nov 2013 | 16:53:25 UTC - in response to Message 34038.

Yes, I have a similar one from the 'short' queue - just approaching 95% after 16 hours on a GTX 470. Looks like it's making slow but steady progress and should complete within the hour - though outside 24 hours from when it was issued.

Perhaps there was a muddle which queue these tasks should be loaded into?

I60-SANTI_baxbimSPW-0-62-RND2323_0

lukeu
Send message
Joined: 14 Oct 11
Posts: 31
Credit: 75,720,504
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34040 - Posted: 27 Nov 2013 | 17:45:44 UTC - in response to Message 34039.

"Me too": http://www.gpugrid.net/workunit.php?wuid=4948159

To abort or not to abort... that is the question.

- Current time: 10h16m
- Estimated remaining: 2h4m, but this number decreases by 1s every 9s clock time.
- When this WU ran on another computer, it exited after 13s with "The simulation has become unstable. Terminating to avoid lock-up (1)"

So my projection is more like "20h remaining" if BOINC's estimate doesn't decelerate further (which I suspect it will).

Yup, I think I just answered my own question: abort. :-)

John
Send message
Joined: 15 Oct 11
Posts: 17
Credit: 81,085,378
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 34041 - Posted: 27 Nov 2013 | 17:54:19 UTC - in response to Message 34039.
Last modified: 27 Nov 2013 | 17:57:33 UTC

%17.75 after twelve hours.
Knocks the crap outa the system.High CPU useage.
Computer basically unusable while crunching.
If I suspend workunit, BOINC?? craps out and I get low ( %5 - %30) GPU usuage and very little progress.After reboot GPU usage between %85 - %90 but still very slow crunching.

Wrong "queue" ??

http://www.gpugrid.net/result.php?resultid=7494659

http://www.gpugrid.net/result.php?resultid=7494640

Stefan
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 5 Mar 13
Posts: 348
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 34042 - Posted: 27 Nov 2013 | 18:50:44 UTC

Looking into it

Profile Stoneageman
Avatar
Send message
Joined: 25 May 09
Posts: 224
Credit: 34,057,224,498
RAC: 231
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34044 - Posted: 27 Nov 2013 | 19:33:36 UTC
Last modified: 27 Nov 2013 | 19:35:24 UTC

These 'short tasks' have all the hallmarks of long tasks except for the credit

Profile dskagcommunity
Avatar
Send message
Joined: 28 Apr 11
Posts: 456
Credit: 817,865,789
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34045 - Posted: 27 Nov 2013 | 20:19:55 UTC

Why i reactiveded short queue -_- got the same..
____________
DSKAG Austria Research Team: http://www.research.dskag.at



Killersocke
Send message
Joined: 18 Oct 13
Posts: 53
Credit: 406,647,419
RAC: 0
Level
Gln
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34050 - Posted: 28 Nov 2013 | 1:06:23 UTC
Last modified: 28 Nov 2013 | 1:12:52 UTC

Puh
Starttime 10:00 o'clock yesterday in the morning
now it's midnight 2:00 o'clock
crunching time round about 16 Hrs !
This was never a short part....
My GPU is so hot I can fry eggs on it

Stefan
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 5 Mar 13
Posts: 348
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 34053 - Posted: 28 Nov 2013 | 8:14:22 UTC

Yes you are right, we apologize a lot. There was a mistake in the sending of the WUs and they were indeed configured for the long queue. Santi has hopefully by now stopped all further WUs (but we won't cancel currently running ones so that people don't lose credit). I assume he will resend today with correct configuration.

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,600,211,851
RAC: 8,778,017
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34055 - Posted: 28 Nov 2013 | 16:36:14 UTC - in response to Message 34053.

Yes you are right, we apologize a lot. There was a mistake in the sending of the WUs and they were indeed configured for the long queue. Santi has hopefully by now stopped all further WUs (but we won't cancel currently running ones so that people don't lose credit). I assume he will resend today with correct configuration.

That seems to have worked - I got new work from the short queue this morning, which looked identical from the outside (SANTI_baxbimSPW2-0-62- sequence, same estimated runtime) but completed in 4 hours instead of 16 hours...

... and was given exactly the same credit!

All tasks for host 43404

Profile dskagcommunity
Avatar
Send message
Joined: 28 Apr 11
Posts: 456
Credit: 817,865,789
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34059 - Posted: 28 Nov 2013 | 18:21:47 UTC - in response to Message 34053.

Yes you are right, we apologize a lot. There was a mistake in the sending of the WUs and they were indeed configured for the long queue. Santi has hopefully by now stopped all further WUs (but we won't cancel currently running ones so that people don't lose credit). I assume he will resend today with correct configuration.


This time i would cancel them..we are talking about 17k credits only what some lose for one unit. Much time lost ok, but i already canceld my unit, because beginning from a fresh long unit give me much more credits then computing the wrong queued unit to the final.
____________
DSKAG Austria Research Team: http://www.research.dskag.at



Stefan
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 5 Mar 13
Posts: 348
Credit: 0
RAC: 0
Level

Scientific publications
wat
Message 34067 - Posted: 29 Nov 2013 | 9:40:11 UTC

Yes we debated it a bit. At this point it's up to the users to cancel them. But considering the complaints last time we accidentally "nuked" them (as we call that type of cancelling now) we thought it better to let it run its hopefully short course and continue with the good WUs.
I don't know how many of the bad ones are still running. We will take a look at it.

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34069 - Posted: 29 Nov 2013 | 11:25:08 UTC - in response to Message 34067.

The WU's are not bad, its just that the credit is poor/wrong.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 1576
Credit: 5,600,211,851
RAC: 8,778,017
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34080 - Posted: 30 Nov 2013 | 0:23:39 UTC

Conversely, the new trypsin_lig_356x4-NOELIA_RC5-0-1- series seems to be running for the same time, but paying almost twice as much credit (same host link as last time).

Swings and roundabouts...

Profile skgiven
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 34084 - Posted: 30 Nov 2013 | 2:04:13 UTC - in response to Message 34080.

I was thinking of the old Hot and Cold taps, where the red and blue colours use to ware off, and in some houses the taps were back to front, so you had to guess which was hot and which was cold.

Maybe the short Trp WU's are being granting credit according to the Long credit system (50% bonus)?
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Post to thread

Message boards : Number crunching : I652-SANTI_baxbimSPW-0-62-RND0982_0

//