Message boards : News : First tests 2013
Message board moderation

To post messages, you must log in.

AuthorMessage
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 843
Credit: 1,578,195
RAC: 0
Message 25073 - Posted: 11 Jan 2013, 12:09:25 UTC

Trying to run a few thousand cases from Scinetific Linux 6 (SLC6)
here at CERN. Eric.
ID: 25073 · Report as offensive     Reply Quote
Profile Robert Pick

Send message
Joined: 1 Dec 05
Posts: 59
Credit: 5,761,949
RAC: 37
Message 25074 - Posted: 11 Jan 2013, 12:36:57 UTC - in response to Message 25073.  

Thank you! Got 24 and they seem to be running as before or just fine!!! Pick
ID: 25074 · Report as offensive     Reply Quote
metalius
Avatar

Send message
Joined: 3 Oct 06
Posts: 99
Credit: 6,361,279
RAC: 51
Message 25075 - Posted: 11 Jan 2013, 12:37:29 UTC - in response to Message 25073.  

Good luck!
ID: 25075 · Report as offensive     Reply Quote
mitrichr
Avatar

Send message
Joined: 20 Dec 07
Posts: 69
Credit: 599,151
RAC: 0
Message 25077 - Posted: 11 Jan 2013, 16:59:38 UTC

Does that mean that the WU's will not run on or go out to Windows?

ID: 25077 · Report as offensive     Reply Quote
Tigis

Send message
Joined: 23 Dec 11
Posts: 3
Credit: 80,747
RAC: 0
Message 25078 - Posted: 11 Jan 2013, 17:18:32 UTC - in response to Message 25077.  

Does that mean that the WU's will not run on or go out to Windows?


I have a Windows XP system and just got 8 tasks
ID: 25078 · Report as offensive     Reply Quote
Richard Haselgrove

Send message
Joined: 27 Oct 07
Posts: 182
Credit: 3,295,818
RAC: 0
Message 25079 - Posted: 11 Jan 2013, 17:23:28 UTC - in response to Message 25078.  

I've been picking them up on Windows too. Some of the short ones have already gone back and validated.
ID: 25079 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 843
Credit: 1,578,195
RAC: 0
Message 25080 - Posted: 11 Jan 2013, 17:30:29 UTC

......actually around 50,000 jobs, BUT many will be very very short.
We are looking at small bunch charges, little beam to beam effect but
at high amplitudes all particles will be lost very quickly! Eric.
ID: 25080 · Report as offensive     Reply Quote
mitrichr
Avatar

Send message
Joined: 20 Dec 07
Posts: 69
Credit: 599,151
RAC: 0
Message 25081 - Posted: 11 Jan 2013, 20:30:17 UTC

Yup, I got some too.

ID: 25081 · Report as offensive     Reply Quote
tullio

Send message
Joined: 19 Feb 08
Posts: 530
Credit: 2,219,362
RAC: 446
Message 25084 - Posted: 12 Jan 2013, 14:02:50 UTC

One task ended in 41 s on my Linux box and validated. The other has been running for 9 hours and is at 55%.
Tullio
ID: 25084 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 843
Credit: 1,578,195
RAC: 0
Message 25098 - Posted: 12 Jan 2013, 20:36:37 UTC

Running well; submitting more cases while I continue to work
on recent problems. I am extending the intensity scan to
higher amplitudes for small bunch charges. Eric.
ID: 25098 · Report as offensive     Reply Quote
Profile ritterm
Avatar

Send message
Joined: 30 May 08
Posts: 88
Credit: 3,840,700
RAC: 3
Message 25099 - Posted: 13 Jan 2013, 2:36:11 UTC

Tasks ready to send 116,621

I love it! :-)
ID: 25099 · Report as offensive     Reply Quote
Profile MAGIC Quantum Mechanic
Avatar

Send message
Joined: 24 Oct 04
Posts: 713
Credit: 21,078,976
RAC: 21,516
Message 25103 - Posted: 13 Jan 2013, 21:53:32 UTC



The credits sure are not very magnetic so far.

I have got several short and long sse2's and the short ones remind me of the old days here (.04 credits per)

And the longer ones taking from 10,000 to 15,000 seconds getting 13 to 23 credits per.

Now if you think credits are no big deal then you won't mind running a computer here but when you have a room full of them and run another project that you like and never stops running then this project will never attract many members.

It would be a real good idea to change those 10,000 to 13,000 second tasks to something like the Einstein GRP's that average around 8000+ seconds and give 337 credits per.

That is the only way to get thousands of computers to help here for the LHC

I have been here since 2004 and that is one thing that is never any better and now it is not as easy to get my 7 hosts running these tasks with the extra cores I have that aren't running the cuda's


Volunteer Mad Scientist For Life
ID: 25103 · Report as offensive     Reply Quote
Richard Haselgrove

Send message
Joined: 27 Oct 07
Posts: 182
Credit: 3,295,818
RAC: 0
Message 25104 - Posted: 13 Jan 2013, 22:32:39 UTC - in response to Message 25103.  

It would be a real good idea to change those 10,000 to 13,000 second tasks to something like the Einstein GRP's that average around 8000+ seconds and give 337 credits per.

I suspect this poster knows very well that the Einstein GRP (Gamma Ray Pulsar) run was divided into two parts.

The original run (which finished before Christmas) awarded 337 credits per task, but typically ran for a great many more than 8,000 seconds.

A new application was deployed over the New Year, with re-designed workunits: those are the ones which run for ~8,000 seconds.

For a temporary period, the new tasks were awarded the same credit as the old ones, while their performance was assessed. The credit per task has now been reduced to 70 - which may indeed be too low for cross-project comparability, and still be subject to review.

I suggest you consult with the project administrators at Einstein before coming to any final decision on this suggestion.
ID: 25104 · Report as offensive     Reply Quote
William C Wilson
Avatar

Send message
Joined: 11 Sep 08
Posts: 25
Credit: 384,225
RAC: 0
Message 25108 - Posted: 14 Jan 2013, 16:15:03 UTC - in response to Message 25104.  

I got 12 task, each ran 4 to 10 seconds and showed 100% completed. Something wrong with them, or just a test? Thank you.
William C Wilson
São Paulo Brazil
ID: 25108 · Report as offensive     Reply Quote
Sunny129
Avatar

Send message
Joined: 12 Dec 05
Posts: 31
Credit: 9,709,398
RAC: 0
Message 25109 - Posted: 14 Jan 2013, 16:26:47 UTC - in response to Message 25108.  

I got 12 task, each ran 4 to 10 seconds and showed 100% completed. Something wrong with them, or just a test? Thank you.

yeah those are normal...its just part of the test run.
ID: 25109 · Report as offensive     Reply Quote
William C Wilson
Avatar

Send message
Joined: 11 Sep 08
Posts: 25
Credit: 384,225
RAC: 0
Message 25110 - Posted: 14 Jan 2013, 18:34:46 UTC - in response to Message 25108.  

The second download of another 14 or so today, about 30 min after the first, are running normal. Did not boot, no changes in machine at all. WEIRD.
William C Wilson
São Paulo Brazil
ID: 25110 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 843
Credit: 1,578,195
RAC: 0
Message 25111 - Posted: 14 Jan 2013, 20:14:15 UTC - in response to Message 25110.  

Not weird; we cannot really predict how long each case will be.
One can guess based on the initial amplitude which can be found
in the task name. We are studying chaos here! Eric.
ID: 25111 · Report as offensive     Reply Quote
Eric Mcintosh
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 12 Jul 11
Posts: 843
Credit: 1,578,195
RAC: 0
Message 25112 - Posted: 14 Jan 2013, 20:14:16 UTC - in response to Message 25110.  

Not weird; we cannot really predict how long each case will be.
One can guess based on the initial amplitude which can be found
in the task name. We are studying chaos here! Eric.
ID: 25112 · Report as offensive     Reply Quote

Message boards : News : First tests 2013


©2018 CERN