| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: FightAIDS@Home Phase 2
Thread: EEK !! 1 x FAAH2 WU per core restriction has been lifted ! Deliberate? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 5
|
|
| Author |
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: 
|
I just noticed that the work caches on those of my devices which run FAAH2 are flooded with FAAH2 WUs, all with 1-day due dates.
The earliest of these that I've seen in a brief farm tour whould have been sent out at around 10:00 UTC 7 June 19. Some of the WUs that I've received won't make the deadline, so I will cull some with the abort button. Surely this can't be a deliberate change. The deadline allowance would have been increased if it was so. |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Maybe a bug in the device profiles where you can choose to have up to 64 if select less than unlimited. Of course, using ncpus in cc_config.xml could do that too, then use app_config to limit the concurrent running.
----------------------------------------[Edit 1 times, last edit by Former Member at Jun 7, 2019 2:33:51 PM] |
||
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges:
|
Thanks, Lavaflow, but because my clients are able to get more than 1 FAHB WU per core, there must have been a change. AFAIK, unannounced.
See Uplinger's post at the time that FAAH Phase II restarted with the new AsyncRE system - https://www.worldcommunitygrid.org/forums/wcg/viewpostinthread?post=553567 Quote: "4. We are limiting results to 1 per core, similar to how beta runs." Setting the absolute number of a specific project's WUs allowed in the cache is a rather blunt instrument when there are so few device profiles available. It doesn't make sense to set the same number for an Intel atom netbook and an AMD 2990WX, which may have to share a common profile :) |
||
|
|
katoda
Senior Cruncher Poland Joined: Apr 28, 2007 Post Count: 170 Status: Offline Project Badges: 
|
A couple days ago I observed this as well, on my 8-thread machine there were around 14 FAAH2 tasks in the queue. Prophylactically I went to device profile settings on the website, checked that "default" limit is still there and saved the profile again. Maybe it is a coincidence or not, but number of FAAH2 units went back to normal after that.
---------------------------------------- |
||
|
|
rod4x4
Cruncher Joined: Apr 29, 2014 Post Count: 12 Status: Offline Project Badges:
|
Same happened here.
----------------------------------------My profile was set to unlimited for FAH2. Changing profile setting to explicit number throttled back the tasks. (after aborting the jobs that could not be completed). Linux FAH2 tasks seemed to have dried up, but still getting FAH2 Windows tasks [Edit 1 times, last edit by rod4x4 at Jun 9, 2019 11:46:55 PM] |
||
|
|
|