Advanced search

Message boards : News : WU: NATHAN_KIDKIX...

Author Message
Profile nate
Send message
Joined: 6 Jun 11
Posts: 124
Credit: 2,928,865
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 32445 - Posted: 29 Aug 2013 | 11:01:31 UTC
Last modified: 29 Aug 2013 | 11:44:37 UTC

Hi all,

Workunits with a similar name have been running for the past month or so on the long queue, but in keeping with our new policy I wanted to create a thread about them here. My original message in a previous thread was:

With these workunits, I'll be looking at the folding of a small peptide using a special force field. The biophysics community uses different force fields to represent the motions of proteins and other molecules, and we are attempting to test a new one so we can check some previous work and feel comfortable in work we are planning to do.


This work is a continuation of that, where we are now looking at the interaction of that peptide with another protein. The idea is to understand the interaction process, and later to see if the lessons we learn extend to other such interactions. There are MANY diseases in which these kinds interactions are important, from many cancers such as prostate cancer to neurological disorders like Parkinson's disease. They are fundamentally important to the function of cells. This is something I'll be looking at in the coming months.

If there are any problems or you have further questions, post them here!

Regards,
Nate

reklov
Send message
Joined: 27 Jan 13
Posts: 3
Credit: 1,016,600
RAC: 0
Level
Ala
Scientific publications
wat
Message 32750 - Posted: 5 Sep 2013 | 21:57:00 UTC - in response to Message 32445.

on my gpu GT620 one of these wu's was downloaded my accident (accept other applications were allowed). Now it already runs 87hours and is only 52% done.
There is no chance to reach the deadline tomorrow morning.
Can the deadline be extended? Or can it be excluded that such long running wu's are downloaded to slow gpus's?
____________

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,470,385,294
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 32753 - Posted: 5 Sep 2013 | 22:03:22 UTC - in response to Message 32750.

What GPU do you have? Is it down clocked by you or by an app?
You can check that, if the GPU core clock runs at half speed or even less, the WU need lots more time to finish.
____________
Greetings from TJ

Profile dskagcommunity
Avatar
Send message
Joined: 28 Apr 11
Posts: 456
Credit: 817,865,789
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 32762 - Posted: 6 Sep 2013 | 0:34:38 UTC

He wrote down already he has a 620gt ;)

Only change your default? profile to short queue only. then the units are shorter too with the same dead,ine. But a 620 card is still a very slow card for this project.
____________
DSKAG Austria Research Team: http://www.research.dskag.at



TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,470,385,294
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 32767 - Posted: 6 Sep 2013 | 8:27:46 UTC - in response to Message 32762.

He wrote down already he has a 620gt ;)

Only change your default? profile to short queue only. then the units are shorter too with the same dead,ine. But a 620 card is still a very slow card for this project.

Indeed you are right. My eyes become old :)
Yeah for this project you need top-cards, otherwise long runs, become extremely long.
____________
Greetings from TJ

reklov
Send message
Joined: 27 Jan 13
Posts: 3
Credit: 1,016,600
RAC: 0
Level
Ala
Scientific publications
wat
Message 32818 - Posted: 7 Sep 2013 | 6:00:10 UTC - in response to Message 32750.

The wu now excessed the deadline as expected, stil about 44 hours to go on my GT620.
If the wu is running to the final end and then returned late, will I get credits on this project? Or is it discarded as too late independent of the outcome of the calculation? I saw already both possible behaviours at other projects.
____________

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 32819 - Posted: 7 Sep 2013 | 7:50:05 UTC - in response to Message 32818.

It will get resent at some stage shortly after the 5th day. If you return the WU before someone else does, I think you may get half credit, but if they return it before you, I think no credit will come your way. I'm not 100% sure on this anymore.

Your system is hidden so I can't see if it's already been resent. It might be worth completing just to see if you do get credit, so others know.
____________
FAQ's

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

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 7,520
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 32822 - Posted: 7 Sep 2013 | 13:53:05 UTC - in response to Message 32819.
Last modified: 7 Sep 2013 | 13:56:19 UTC

It's already in the timed out - no response state, but it's not resent yet to another host, but it will be soon. Reklov will receive credit only, when Reklov's host returns the result before the 3rd host does. In this case the the 3rd host will loose the 24h or 48h credit bonus, and both hosts will receive only the standard credit amount.

reklov
Send message
Joined: 27 Jan 13
Posts: 3
Credit: 1,016,600
RAC: 0
Level
Ala
Scientific publications
wat
Message 32855 - Posted: 9 Sep 2013 | 5:53:27 UTC - in response to Message 32822.

The wu finished this morning before the resent got out. It validated and got it's credits.
____________

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 32858 - Posted: 9 Sep 2013 | 7:34:18 UTC - in response to Message 32855.

Excellent! Glad you got the credits, but you were lucky.
I suggest you stick to the short WU's - it's really not nice to miss the deadline and not get credits (been there a few times for various reasons).
____________
FAQ's

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

werdwerdus
Send message
Joined: 15 Apr 10
Posts: 123
Credit: 1,004,473,861
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 32882 - Posted: 11 Sep 2013 | 7:57:58 UTC

I54R3-NATHAN_KIDc22_glu-0-10-RND8113_0

windows 7 gtx 660 ti, using 83% gpu, a little low here
____________
XtremeSystems.org - #1 Team in GPUGrid

Jacob Klein
Send message
Joined: 11 Oct 08
Posts: 1127
Credit: 1,901,927,545
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 33325 - Posted: 2 Oct 2013 | 10:23:26 UTC - in response to Message 32445.
Last modified: 2 Oct 2013 | 10:35:33 UTC

These work units really give my video cards a workout (which is a good thing).

When no other CPU tasks are running, I am seeing:
eVGA GTX 460 OC 1GB: 96% GPU Utilization, moderate-high temps (69*C at 82% GPU fan)
eVGA GTX 660 Ti FTW 3GB: 95% GPU Utilization, very-high temps (72*C at 80% [the max] GPU fan)

I have not seen any other work unit type work my GPUs this hard. And the 6.14 app (which fixes my suspend/resume scenarios) ensures the work gets processed without error.

I hope you get the results you are after!

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,470,385,294
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 33328 - Posted: 2 Oct 2013 | 15:20:00 UTC - in response to Message 33325.

I have one running at 94% load at 64°C, on a EVGA 660, 2GB, not OC, 1048MHz, Win7 professional 64 bit.
I have had a Noelia earlier this week, running at 96% load.
____________
Greetings from TJ

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 33415 - Posted: 8 Oct 2013 | 9:42:20 UTC - in response to Message 33328.
Last modified: 8 Oct 2013 | 11:57:24 UTC

A NATHAN_KIDKIXc22 WU 'dozed off' on my GTX650TiBoost; work was not progressing and the GPU was cold...
I suspended the WU, and the SANTI_RAP74wtCUBIC WU running on it's neighboring (GTX670, GPU0), then resumed the NATHAN WU, which made it run on GPU0. Both are now progressing from their last checkpoints.

I12R5-NATHAN_KIDKIXc22_6-15-50-RND4167_0... Completed and Validated.
____________
FAQ's

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

Jeremy Zimmerman
Send message
Joined: 13 Apr 13
Posts: 61
Credit: 726,605,417
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwat
Message 33457 - Posted: 11 Oct 2013 | 10:32:33 UTC

Looks like a bad Nathan WU (I84R5-NATHAN_KIDKIXc22_6-16-50-RND6324)
http://www.gpugrid.net/workunit.php?wuid=4825390

Neither Cuda 42 or 55 could run this one.
Nor could...
# GPU [GeForce GTX 680] Platform [Windows] Rev [3203] VERSION [42]
Or
# GPU [GeForce GTX 690] Platform [Windows] Rev [3203] VERSION [55]
Or
# GPU [GeForce GTX 660 Ti] Platform [Windows] Rev [3203] VERSION [55]
Or
# GPU [GeForce GTX 660 Ti] Platform [Windows] Rev [3203] VERSION [42]
Or
# GPU [GeForce GTX 680] Platform [Windows] Rev [3203] VERSION [55]

All gave the following error within 10 seconds.
(unknown error) - exit code -97 (0xffffff9f)
# The simulation has become unstable. Terminating to avoid lock-up (1)

Post to thread

Message boards : News : WU: NATHAN_KIDKIX...

//