is this pausing and restarting normal?

Mike Box
Mike Box
Joined: 13 Jul 06
Posts: 3
Credit: 2780
RAC: 0
Topic 191579

14/07/2006 13:59:01|Einstein@Home|URL: http://einstein.phys.uwm.edu/; Computer ID: 691430; location: ; project prefs: default
14/07/2006 13:59:01|Einstein@Home|Deferring task l1_1047.0_S5R1__460_S5R1a_1
14/07/2006 19:54:53|Einstein@Home|Restarting task l1_1047.0_S5R1__460_S5R1a_1 using einstein_S5R1 version 402
14/07/2006 21:54:54|Einstein@Home|Pausing task l1_1047.0_S5R1__460_S5R1a_1 (removed from memory)
15/07/2006 02:29:29|Einstein@Home|Restarting task l1_1047.0_S5R1__460_S5R1a_1 using einstein_S5R1 version 402
15/07/2006 04:29:29|Einstein@Home|Pausing task l1_1047.0_S5R1__460_S5R1a_1 (removed from memory)
15/07/2006 10:42:58|Einstein@Home|Restarting task l1_1047.0_S5R1__460_S5R1a_1 using einstein_S5R1 version 402
15/07/2006 11:23:40|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
15/07/2006 11:23:40|Einstein@Home|Reason: To fetch work
15/07/2006 11:23:40|Einstein@Home|Requesting 6 seconds of new work
15/07/2006 11:23:45|Einstein@Home|Scheduler request succeeded
15/07/2006 11:23:47|Einstein@Home|Pausing task l1_1047.0_S5R1__460_S5R1a_1 (removed from memory)
15/07/2006 12:34:28|Einstein@Home|Restarting task l1_1047.0_S5R1__460_S5R1a_1 using einstein_S5R1 version 402
15/07/2006 13:11:34|Einstein@Home|Computation for task l1_1047.0_S5R1__460_S5R1a_1 finished
15/07/2006 13:11:36|Einstein@Home|Started upload of file l1_1047.0_S5R1__460_S5R1a_1_0
15/07/2006 13:11:41|Einstein@Home|Finished upload of file l1_1047.0_S5R1__460_S5R1a_1_0
15/07/2006 13:11:41|Einstein@Home|Throughput 59591 bytes/sec
15/07/2006 18:18:28|Einstein@Home|Starting task l1_1047.0_S5R1__457_S5R1a_1 using einstein_S5R1 version 402
15/07/2006 20:18:29|Einstein@Home|Pausing task l1_1047.0_S5R1__457_S5R1a_1 (removed from memory)
16/07/2006 01:54:36|Einstein@Home|Restarting task l1_1047.0_S5R1__457_S5R1a_1 using einstein_S5R1 version 402
16/07/2006 03:54:56|Einstein@Home|Pausing task l1_1047.0_S5R1__457_S5R1a_1 (removed from memory)
16/07/2006 09:39:37|Einstein@Home|Restarting task l1_1047.0_S5R1__457_S5R1a_1 using einstein_S5R1 version 402
16/07/2006 11:39:38|Einstein@Home|Pausing task l1_1047.0_S5R1__457_S5R1a_1 (removed from memory)
16/07/2006 18:15:59|Einstein@Home|Restarting task l1_1047.0_S5R1__457_S5R1a_1 using einstein_S5R1 version 402
16/07/2006 20:16:00|Einstein@Home|Pausing task l1_1047.0_S5R1__457_S5R1a_1 (removed from memory)
17/07/2006 01:17:54|Einstein@Home|Restarting task l1_1047.0_S5R1__457_S5R1a_1 using einstein_S5R1 version 402
17/07/2006 03:17:55|Einstein@Home|Pausing task l1_1047.0_S5R1__457_S5R1a_1 (removed from memory)

anders n
anders n
Joined: 29 Aug 05
Posts: 123
Credit: 1656300
RAC: 0

is this pausing and restarting normal?

Hi Mike

Are you crunching for other projects as well?

And if so what are your resource share settings for the

projects?

Anders n

Mike Box
Mike Box
Joined: 13 Jul 06
Posts: 3
Credit: 2780
RAC: 0

Hi Anders, Thanks for your

Hi Anders,

Thanks for your rapid response

I'm helping to crunch 4 projects - BBC Climate Change, Rosetta , Malaria and Einstein. All have 25% resource allocations.

Mike

anders n
anders n
Joined: 29 Aug 05
Posts: 123
Credit: 1656300
RAC: 0

Then it seems to work just

Then it seems to work just fine.

You could try out setting "leave in memory" to yes as you loose some

cruching every time a WU is preempted.

Anders n

Winterknight
Winterknight
Joined: 4 Jun 05
Posts: 1253
Credit: 323765635
RAC: 386828

BOINC switches between

BOINC switches between projects at the period set in 'your account' general preferences, the default is 60 mins.
It switches to enable all projects to get a fair share, as you can see not all projects have equal length units and even in Eistein there are long and short units. If it did not switch and you started the Climate change unit, probably several 100's of hours, the other project units would miss their deadlines.

BOINC will also check when you download a new unit, and possibly switch units, this is so that any very short units with tight margins are crunched asap.

What project is crunched and which project next gets a download is based on a system of debts. Short term (STD) to determine which is crunched and long term (LTD) for the downloads. In each case the highest positive value is used, unless a unit will miss its deadline if not crunched soon, this known as Earliest Deadline first (EDF), in which case it will run up a high STD and allow the other projects to catch up their fair share.

You can get more information from the Unofficial Wiki
Andy

Mike Box
Mike Box
Joined: 13 Jul 06
Posts: 3
Credit: 2780
RAC: 0

Thanks for the info. My

Thanks for the info. My general preferences have now been changed to keeping applications in memory and I'm getting a better understanding of how things work.

m

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.