Message boards : Number crunching : Available Workunits
Message board moderation

To post messages, you must log in.

AuthorMessage
Mark W. Patton

Send message
Joined: 2 Sep 09
Posts: 5
Credit: 877,912
RAC: 0
Message 24708 - Posted: 24 Aug 2012, 19:18:14 UTC

I am really excited to be doing workunits now for the largest scientific instrument on the planet. I am trying to reconfigure the machines I have to get more work out of them. Keep up the good work. I am sure I have no idea how difficult it can be to operate such a machine. I hope in the near future to upgrade my computing capabilities to be able to do more of this important work. Try to ignore impatient people and realise there are some of us out here that appreciate your efforts.
Regards,
Mark W. Patton
dracona7@yahoo.com
ID: 24708 · Report as offensive     Reply Quote
Michael Karlinsky
Avatar

Send message
Joined: 18 Sep 04
Posts: 163
Credit: 1,682,370
RAC: 0
Message 24709 - Posted: 24 Aug 2012, 20:05:47 UTC - in response to Message 24708.  
Last modified: 24 Aug 2012, 20:25:16 UTC

I am really excited to be doing workunits now for the largest scientific instrument on the planet. I am trying to reconfigure the machines I have to get more work out of them. Keep up the good work. I am sure I have no idea how difficult it can be to operate such a machine. I hope in the near future to upgrade my computing capabilities to be able to do more of this important work. Try to ignore impatient people and realise there are some of us out here that appreciate your efforts.
Regards,
Mark W. Patton
dracona7@yahoo.com


Yes, we *never* had that much available WUs. 450+K is way more than in the good old days of 2004. As far as I remember 100K was the average.

Michael
Team Linux Users Everywhere
ID: 24709 · Report as offensive     Reply Quote
Zonar

Send message
Joined: 14 Jul 12
Posts: 2
Credit: 173,287
RAC: 0
Message 24719 - Posted: 25 Aug 2012, 15:55:57 UTC

It's nice to see that there are a lot of work units available now for us, but I have one slight concern: When a new task is created because the validation of a work unit is inconclusive (http://lhcathomeclassic.cern.ch/sixtrack/workunit.php?wuid=2798195) this task is inserted in the middle of the queue so it would take days before it is assigned to a host. Is it possible to insert these tasks right on the front of the queue, like what happens in SIMAP?
ID: 24719 · Report as offensive     Reply Quote
Profile Tom95134

Send message
Joined: 4 May 07
Posts: 250
Credit: 826,541
RAC: 0
Message 24721 - Posted: 25 Aug 2012, 17:19:56 UTC - in response to Message 24719.  

It's nice to see that there are a lot of work units available now for us, but I have one slight concern: When a new task is created because the validation of a work unit is inconclusive (http://lhcathomeclassic.cern.ch/sixtrack/workunit.php?wuid=2798195) this task is inserted in the middle of the queue so it would take days before it is assigned to a host. Is it possible to insert these tasks right on the front of the queue, like what happens in SIMAP?

With the large number of tasks it would be nice if we're waiting for validation of one we've run and it has bombed on the partner to not have to wait so long but...

I think the concern about putting it at the front of the queue would be the possibility of it getting pushed to the same client. My observation indicates that Task uploading and reporting happens first and then the client may or may not request new Tasks. If the one that bombed gets put at the front of the queue it might just get pushed to the same client and bomb again. If you bomb tasks enough times then you don't get new Tasks.

Of course that's just my observation since I have no knowledge about what is happening inside the servers.
ID: 24721 · Report as offensive     Reply Quote

Message boards : Number crunching : Available Workunits


©2024 CERN