New longrunners


Advanced search

Message boards : Number crunching : New longrunners

Author Message
Profile Yeti
Avatar
Send message
Joined: 20 Jul 14
Posts: 699
Credit: 22,597,832
RAC: 0
    
Message 6162 - Posted: 21 Feb 2017, 13:37:20 UTC
Last modified: 21 Feb 2017, 13:38:49 UTC

@Projectteam:

I have three longrunners at my network, that are something 30x the normal run-time (e.g. 300 hours instead of 10 hours), but they all look like they are still really running:


CPU-Power is consumed

SingleCoreWU:
CPU-Time is 300 hours
Elapsed Time is 300 hours

MultiCoreWU 4c:
CPU-Time: 38 days
Elapsed Time is 10 days

Kepp them running or abort ?

Note, the SingleCore-WUs are short before final Deadline

David Cameron
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 13 May 14
Posts: 252
Credit: 2,028,556
RAC: 1
    
Message 6163 - Posted: 21 Feb 2017, 14:23:57 UTC - in response to Message 6162.

Hi Yeti,

Can you abort one of them and I will take a look at the log?

Profile Yeti
Avatar
Send message
Joined: 20 Jul 14
Posts: 699
Credit: 22,597,832
RAC: 0
    
Message 6164 - Posted: 21 Feb 2017, 14:27:02 UTC

Here a MultiCoreWU:

http://atlasathome.cern.ch/result.php?resultid=8309479

David Cameron
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 13 May 14
Posts: 252
Credit: 2,028,556
RAC: 1
    
Message 6167 - Posted: 22 Feb 2017, 9:10:36 UTC - in response to Message 6164.

Unfortunately I wasn't quick enough and the cleaner script deleted the output before I could get to it...

Can you abort another one and send me a private message so I can look at it straight away?

James Martin
Send message
Joined: 12 Jul 14
Posts: 4
Credit: 24,458
RAC: 0
    
Message 6283 - Posted: 1 Apr 2017, 0:36:28 UTC

Backing off 03:45:11 on upload of XvzMDm2seDqn7jp7oou28CBqABFKDmABFKDm27FKDmYdyKDm7EpOTo_0_r1042369581_ATLAS_result

Atlas simulation running on multiple core_1.04 (vbox 64...)


The above message was one of a number of attempts to upload the WU. The WU's
size is 53.85 MB. The upload failures result in retries of approx. 3hrs30+ mins.

My computer is a Dell Latitude E7240, w/4 CPU's.

Will this help? Need more info.?

James Martin
Send message
Joined: 12 Jul 14
Posts: 4
Credit: 24,458
RAC: 0
    
Message 6284 - Posted: 1 Apr 2017, 0:48:30 UTC

Backing off 03:45:11 on upload of XvzMDm2seDqn7jp7oou28CBqABFKDmABFKDm27FKDmYdyKDm7EpOTo_0_r1042369581_ATLAS_result

Atlas simulation running on multiple core_1.04 (vbox 64...)


The above message was one of a number of attempts to upload the WU. The WU's
size is 53.85 MB. The upload failures result in retries of approx. 3hrs30+ mins.

My computer is a Dell Latitude E7240, w/4 CPU's.

Will this help? Need more info.?

PHILIPPE
Send message
Joined: 24 Jul 16
Posts: 84
Credit: 53,413
RAC: 0
    
Message 6285 - Posted: 1 Apr 2017, 10:34:32 UTC - in response to Message 6284.

Hi , James Martin ,

sometimes it happens that upload fails.
David Cameron has to clean up the previous upload incomplete on the server side in order to enable a second upload.
So , don't worry.
Your job isn't lost.You have to wait untill he does.But this is week-end so it may happens Monday only.

James Martin
Send message
Joined: 12 Jul 14
Posts: 4
Credit: 24,458
RAC: 0
    
Message 6286 - Posted: 1 Apr 2017, 19:18:57 UTC - in response to Message 6285.

Thanks, Philippe, for the info. I hope uploading is not dependent upon, now,
running other programs. Am going to download a few other, non-Atlas WU's,
after a second, Atlas, extra-long-runner has completed.

James Martin
Send message
Joined: 12 Jul 14
Posts: 4
Credit: 24,458
RAC: 0
    
Message 6287 - Posted: 4 Apr 2017, 1:36:50 UTC - in response to Message 6286.

Philippe -- As a follow-up, the delayed WU, finally, has been successfully
uploaded. Again, thanks, for the heads-up.

PHILIPPE
Send message
Joined: 24 Jul 16
Posts: 84
Credit: 53,413
RAC: 0
    
Message 6289 - Posted: 4 Apr 2017, 16:42:33 UTC - in response to Message 6287.
Last modified: 4 Apr 2017, 16:43:04 UTC

Thanks for the feedback , James , but i think David Cameron deserves also some thanks because he has to manage 2 sites Atlas and LHC at the same times during the consolidation.It's not obvious to keep the eyes at the right place , at the good time.
It stays yet 363 single cores and 836 multi cores wus in progress...
The 7 april , all the remaining tasks will be cancelled.
When this date arrives , everyone is invited to join LHC project , in order to continue this crunching experience.

Add this new project to boinc manager , and at the prompt ,write :

https://lhcathome.cern.ch/lhcathome/

Have a nice crunch...

Message boards : Number crunching : New longrunners