Advanced search

Message boards : Number crunching : ala_structure_81-NOELIA_sh2forces-0-5-RND6951_0

Author Message
valterc
Send message
Joined: 21 Jun 10
Posts: 21
Credit: 6,180,609,672
RAC: 2,195,286
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27583 - Posted: 6 Dec 2012 | 10:03:06 UTC

Hi all, I just suspended this task after running for ~9 hours, other > 11 expected on my GTX570 (should I expect ~300.000 credits for this?). First one of this type that I got. While running it used about 10-13% gpu...

Any comments?

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1112
Credit: 6,162,416,256
RAC: 0
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27584 - Posted: 6 Dec 2012 | 13:01:04 UTC - in response to Message 27583.

First one of this type that I got. While running it used about 10-13% gpu...

Any comments?

The previous NOELIA WUs ran fine for me except for the one I got yesterday:

ala_structure_571-NOELIA_sh2forces-0-5-RND2523

It ran SLOW and only used 13% GPU instead of the usual 90%+. Aborted :-(
I'll wager that something is wrong with these.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 851
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27587 - Posted: 6 Dec 2012 | 13:58:17 UTC - in response to Message 27584.

+1

It was running for more than 5 hours, even so the running time is 0 according to the task's page.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 851
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27588 - Posted: 6 Dec 2012 | 14:02:38 UTC

Here is another one. (over 13h)

Profile dskagcommunity
Avatar
Send message
Joined: 28 Apr 11
Posts: 456
Credit: 817,865,789
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27590 - Posted: 6 Dec 2012 | 16:20:53 UTC

omg i got one too -_- will see tomorrow how far it is :/ tomorrow i build in a new graka...hope i can see how fast it is..not waiting task with 15% gpu load until it finished O.o
____________
DSKAG Austria Research Team: http://www.research.dskag.at



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 27592 - Posted: 6 Dec 2012 | 17:25:59 UTC - in response to Message 27590.

The last NOELIA task I ran was probably from a different batch,
ptyr_structure_471_bis-NOELIA_sh2NOTCL-2-3-RND8206_0 3905816 139859 3 Dec 2012 | 3:50:05 UTC 3 Dec 2012 | 17:58:33 UTC Completed and validated 50,105.85 7,051.59 102,900.00 Long runs (8-12 hours on fastest card) v6.16 (cuda42)

Perhaps there is an issue with this batch?

If anyone is running one of these tasks, does freeing up more CPU cores/threads help?
____________
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: 851
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27593 - Posted: 6 Dec 2012 | 17:42:25 UTC - in response to Message 27592.

The last NOELIA task I ran was probably from a different batch,
ptyr_structure_471_bis-NOELIA_sh2NOTCL-2-3-RND8206_0 Completed and validated

This is a different batch, I've crunched a couple of them without any problems.

Perhaps there is an issue with this batch?

That's sure!

If anyone is running one of these tasks, does freeing up more CPU cores/threads help?

No.

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 27595 - Posted: 6 Dec 2012 | 18:04:14 UTC - in response to Message 27593.

Thanks, makes it more likely that it's a bad batch.
Does anything else about these task grab you as being odd; extra-large amounts of system RAM or GDDR5, memory controller load...?
____________
FAQ's

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

mwgiii
Send message
Joined: 22 Jan 09
Posts: 8
Credit: 988,332,833
RAC: 12,933
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27605 - Posted: 7 Dec 2012 | 15:47:00 UTC - in response to Message 27595.

I had this one the previous two days:
ala_structure_551-NOELIA_sh2forces-0-5-RND0004_0

it validated but run time was 119,668.62 while cpu time was 107,060.00.

Output:
Stderr output

<core_client_version>7.0.28</core_client_version>
<![CDATA[
<stderr_txt>
MDIO: cannot open file "restart.coor"
# Time per step (avg over 2000000 steps): 59.854 ms
# Approximate elapsed time for entire WU: 119708.496 s
called boinc_finish

</stderr_txt>
]]>

I only noticed it was running because my cruncher was so quiet. I thought my GPU fan might have died. I checked EVGA Precision, GPU usage was 7%. I was already over 30 hours crunching the wu on my 570GTX so I let it finish.

I have picked up another one this morning. I'm a little over 1 hour in and at 3% complete with 7% GPU usage.
____________

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 2343
Credit: 16,201,255,749
RAC: 851
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27607 - Posted: 7 Dec 2012 | 18:54:29 UTC - in response to Message 27595.

Does anything else about these task grab you as being odd; extra-large amounts of system RAM or GDDR5, memory controller load...?

Memory controller load: 1-2%

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 27610 - Posted: 7 Dec 2012 | 20:16:36 UTC - in response to Message 27607.
Last modified: 7 Dec 2012 | 20:34:47 UTC

That's far too low!
It suggests the task isn't using the GPU memory efficiently, and is probably doing something on the CPU slowly and not multi-threaded.
You could probably use process explorer to find out what.

I see it timed out on an NVIDIA GeForce 310M. It tried to run using cuda31 - No chance.
____________
FAQ's

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

Profile dskagcommunity
Avatar
Send message
Joined: 28 Apr 11
Posts: 456
Credit: 817,865,789
RAC: 0
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27636 - Posted: 10 Dec 2012 | 13:39:43 UTC

Oh no finally i found a noelia workunit myself with 13% gpu load (3% memory load) after 15hours :/ I will not abort it, but.. :(
____________
DSKAG Austria Research Team: http://www.research.dskag.at



zablociak
Send message
Joined: 2 Jul 12
Posts: 9
Credit: 302,966,028
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwat
Message 27638 - Posted: 10 Dec 2012 | 17:03:29 UTC

I had to kill two of those WUs in last two days.

Not cool.

werdwerdus
Send message
Joined: 15 Apr 10
Posts: 123
Credit: 1,004,473,861
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27646 - Posted: 11 Dec 2012 | 3:17:51 UTC

i'm aborting when I notice them.

get around 10-13% load on winxp gtx 660 Ti and win7 gtx 470
____________
XtremeSystems.org - #1 Team in GPUGrid

[AF>Belgique] bill1170
Send message
Joined: 4 Jan 09
Posts: 13
Credit: 835,602,199
RAC: 80,834
Level
Glu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27654 - Posted: 11 Dec 2012 | 17:03:11 UTC - in response to Message 27646.
Last modified: 11 Dec 2012 | 17:08:15 UTC

Same problem with this one :

http://www.gpugrid.net/workunit.php?wuid=3934567

5-7% gpu load. 2% memory controller load on GTX660Ti

I will abort it. But instructions from the scientist would help :-)

[edit] I got a second one, same problem. I'm moving to short queue until the question is solved of fixed.

Dylan
Send message
Joined: 16 Jul 12
Posts: 98
Credit: 386,043,752
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwat
Message 27664 - Posted: 12 Dec 2012 | 1:13:19 UTC

After running one of these tasks for 24 hours straight, I had to abort it. Sorry. My gpu usage on a 670 was about 8-12%.

Profile Chilean
Avatar
Send message
Joined: 8 Oct 12
Posts: 98
Credit: 385,652,461
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27665 - Posted: 12 Dec 2012 | 2:17:11 UTC

I just aborted a NOELIA, it was using 1-2% GPU. It took it over an hour to get to 2.120% progress.

jlhal
Send message
Joined: 1 Mar 10
Posts: 147
Credit: 1,077,535,540
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27668 - Posted: 12 Dec 2012 | 12:41:30 UTC - in response to Message 27665.
Last modified: 12 Dec 2012 | 12:42:20 UTC

Hi !
Same for me on a GTX690 ! More than 24 hours and still running !

http://www.gpugrid.net/result.php?resultid=6180059

This is the 3rd WU of this kind and I already cancelled previous 2 .

Any action from the author of this "snail" NOELIA batch ?

How can he be signaled ? Is he aware of this ?
____________
Lubuntu 16.04.1 LTS x64

jlhal
Send message
Joined: 1 Mar 10
Posts: 147
Credit: 1,077,535,540
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 27675 - Posted: 12 Dec 2012 | 17:03:39 UTC - in response to Message 27668.

Any action from the author of this "snail" NOELIA batch ?

How can he be signaled ? Is he aware of this ?


The answer is there http://www.gpugrid.net/forum_thread.php?id=3158&nowrap=true#27672

Thanks Noelia !
____________
Lubuntu 16.04.1 LTS x64

Post to thread

Message boards : Number crunching : ala_structure_81-NOELIA_sh2forces-0-5-RND6951_0

//