| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: GPU Support Forum
Thread: GPU Queue |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 17
|
|
| Author |
|
|
ryan222h
Senior Cruncher Joined: Sep 4, 2006 Post Count: 425 Status: Offline |
Are there any plans to increase the workunits available for queue? I currently seem to be limited to about 3 hours of GPU work, which is a little unsettling should an outage occur. My minimum work buffer is set to 2 days.
----------------------------------------10/25/2012 8:08:12 AM | World Community Grid | Requesting new tasks for ATI 10/25/2012 8:08:15 AM | World Community Grid | Scheduler request completed: got 0 new tasks 10/25/2012 8:08:15 AM | World Community Grid | No tasks sent 10/25/2012 8:08:15 AM | World Community Grid | No tasks are available for Help Conquer Cancer 10/25/2012 8:08:15 AM | World Community Grid | This computer has reached a limit on tasks in progress 10/25/2012 8:08:15 AM | World Community Grid | Project has no tasks available  |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
The limit is 100 tasks per GPU in cache no matter what your settings above that limit may be.
----------------------------------------
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
See [post knreed. The techs are working on packaging 2 HCC WU's in one task. Whilst unsettling to some hi-throughput crunchers, this it how it is... the feeder lines are able to keep up in the present balance with 1.5 million per day. When content per HCC WU is doubled, that would then turn to 6 hours of caching for a GPU cruncher.
|
||
|
|
ryan222h
Senior Cruncher Joined: Sep 4, 2006 Post Count: 425 Status: Offline |
That will help some...24 hours would be nicer but would result in a huge number of tasks in queue I suppose.
---------------------------------------- |
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges: 
|
Maybe combine 3 to help the servers out more.
----------------------------------------
Crunching for humanity since 2007!
 |
||
|
|
ryan222h
Senior Cruncher Joined: Sep 4, 2006 Post Count: 425 Status: Offline |
To the techs:
----------------------------------------Any word on increasing cache limits of GPU workunits, or packaging multiples into one workunit? I hear some users ran dry today because of the low cache availability. Thanks |
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges:
|
To the techs: Any word on increasing cache limits of GPU workunits, or packaging multiples into one workunit? I hear some users ran dry today because of the low cache availability. Thanks What he said!  I lost about 4 hours today.
Crunching for humanity since 2007!
----------------------------------------[Edit 1 times, last edit by Bearcat at Nov 5, 2012 8:08:44 PM] |
||
|
|
mmstick
Senior Cruncher Joined: Aug 19, 2010 Post Count: 151 Status: Offline Project Badges: 
|
Same here, I lost many hours since I can compute 100 work units in under 40 minutes with my 7950. 2 work units in a GPU task sounds too pansy, at least pack 50 work units in one, or 100.
----------------------------------------[Edit 1 times, last edit by mmstick at Nov 6, 2012 7:02:41 AM] |
||
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges: 
|
but but....what about us guys who can only do two gpu workunits in 40 mins?
With any move of my mouse, i lose whatever work has been done by my puny gpu. Packing 100 units would mean 30 hours work for me....way more of a cache than I would want. and how about enlarged filesizes and the issues that raises. I am sure that the techs chose double units carefully. |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
We increased the following limits:
max gpu jobs in progress: 200 max cpu jobs in progress: 40 max gpu jobs per day: 4500 max cpu jobs per day: 300 max jobs per scheduling request: 15 As far as packaging multiple images per job. We will only do 2 images per job at this time (once we release the new app versions) due to BOINC not having a way to handle jobs of varying sizes. In the future we hope to work with BOINC so that we can offer a range of jobs so that powerful machines can get big jobs and less powerful machines can get smaller jobs. |
||
|
|
|