rpi_logo
New N-Body
New N-Body
log in

Advanced search

Message boards : News : New N-Body

Author Message
Jake Bauer
Project developer
Project tester
Project scientist
Send message
Joined: 20 Aug 12
Posts: 66
Credit: 406,916
RAC: 0

Message 57086 - Posted: 30 Jan 2013, 17:27:43 UTC

Hello everyone,

I have posted a new run for N-Body. Post feedback here.

Thanks,

Jake

EdwardPF
Send message
Joined: 8 Apr 10
Posts: 25
Credit: 268,525
RAC: 0

Message 57102 - Posted: 31 Jan 2013, 16:31:26 UTC - in response to Message 57086.

All of my n-body WU are erroring out with a DLL error (can't find entry point - I believe)

Ed F

CraigGibbard
Send message
Joined: 23 Jul 10
Posts: 1
Credit: 746,446
RAC: 189

Message 57107 - Posted: 31 Jan 2013, 19:57:37 UTC

Excessively long completion time.

Workunit name de_nbody_100K_1_30_13_1358941502_18929
E computation size 1012820000 GFLOPS
ET Remaining 23255:40:00

I think this might be a small issue.

Craig[/img]

EdwardPF
Send message
Joined: 8 Apr 10
Posts: 25
Credit: 268,525
RAC: 0

Message 57114 - Posted: 31 Jan 2013, 23:34:03 UTC
Last modified: 31 Jan 2013, 23:35:02 UTC

o.k. I continued to get

the procedure entry point pthread_attr_destroy could not be
located in the dynamic link library pthreadGC2_64.dll


so I reset, removed, and reconnected to the project.

I set pref's to n-body only and

" ... use at most 12.50% of the processors"

The n-body appears to be running fine with a est time of 56Hrs.

The 1 WU is consuming 46% of the system CPU resources.

Ed F

win-7 (64 bit core-7 "8" cpu's)

EdwardPF
Send message
Joined: 8 Apr 10
Posts: 25
Credit: 268,525
RAC: 0

Message 57115 - Posted: 1 Feb 2013, 1:32:16 UTC - in response to Message 57114.

well at 2:00:00 the WU is 2.606% done ... I think that implies about 76.75Hours total time.

I ASSUME ;-) it will bomb out before then ( disk space exceeded) ... but we'll see ...

Ed F

Jeffery M. Thompson
Volunteer moderator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 23 Sep 12
Posts: 152
Credit: 12,997,107
RAC: 3

Message 57117 - Posted: 1 Feb 2013, 2:44:10 UTC

We are sweeping a wide range of parameters and I have been seeing really small units and massive units from the combination of parameters.

We are seeing some 12 hour estimates running in minutes. So these units seem to be doing otherwise.

Can you post the work unit name as in Craig's post so I can use the parameters as tests in the future versions.

I can pull test points from the database but having these descriptions with the parameters makes it easy to use as a test case in resolving the issue.

Jeff

Profile Werkstatt
Send message
Joined: 19 Feb 08
Posts: 350
Credit: 124,256,054
RAC: 7,241

Message 57120 - Posted: 1 Feb 2013, 6:49:39 UTC

I have 2 not validated wu's, too many errors. The same for the wingmen.
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=303695888
http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=303864584

Profile Michael Yusko
Avatar
Send message
Joined: 27 May 11
Posts: 22
Credit: 74,494,481
RAC: 195,385

Message 57121 - Posted: 1 Feb 2013, 7:29:57 UTC

Linux64: Utilizing ".05 CPUs + 1 ATI GPU" in reality, a single nbody_100k task is using 156% cpu (1 task is maxing out both CPU cores) Granted they are only taking 2 - 20 minutes each to process, but they are not utilizing the GPU at all.
____________
-mpyusko

AMD FX-8350 @ 4.3GHz
AMD Radeon RX 480 8GB @ 1342MHz/2000MHz

Jeffery M. Thompson
Volunteer moderator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 23 Sep 12
Posts: 152
Credit: 12,997,107
RAC: 3

Message 57122 - Posted: 1 Feb 2013, 7:44:51 UTC - in response to Message 57120.

If they have the error - exit code -1073741511 (0xc0000139)
They may need to reset the project. A copy of the dlls messed up and the files had to be replaced in the database and download directory. That has been working on the windows 7 clients. As for windows 8 I have not had a test on that system yet so if it persists on the Windows 8 clients after a reset please let me know.

Jeff

EdwardPF
Send message
Joined: 8 Apr 10
Posts: 25
Credit: 268,525
RAC: 0

Message 57135 - Posted: 1 Feb 2013, 16:12:04 UTC - in response to Message 57115.

With the WU 18.9% done and 16.5 hours into it ... it has consumed 59 hours of cpu time to date. this is a ratio of about 3.5cpu hrs to 1 wall hour. Some of this is due to computer usage no doubt but the remainder of the lost half hour I assume is due to thread synchronization. Is that a reasonable assumption?

Ed F

EdwardPF
Send message
Joined: 8 Apr 10
Posts: 25
Credit: 268,525
RAC: 0

Message 57175 - Posted: 5 Feb 2013, 21:03:52 UTC - in response to Message 57135.

The WU finished with status "success".

Wall time: 379,018 sec
CPU time: 1,315,527 sec CT/WT = 3.471 (I was using 4 CPU's)
credit: 3004. 9

If I was running the standard MW WU's this much CPU time would have yielded about 33000 Cobblestones (o.k. maybe 31000 would be closer ... ??)

But the thing worked!!

Ed F


Post to thread

Message boards : News : New N-Body


Main page · Your account · Message boards


Copyright © 2018 AstroInformatics Group