| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: Questions about participation drive for CEP2 |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 152
|
|
| Author |
|
|
Mamajuanauk
Master Cruncher United Kingdom Joined: Dec 15, 2012 Post Count: 1900 Status: Offline Project Badges: 
|
There is significant differences in the lenght of CEP2 wu's, I have some scheduled for around 8 hours, others are listed as much as 14 hours, others anywhere in between...
----------------------------------------ALthough, it's worth mentioning that the 14 hour wu's are on the 2 slower machines, the rest are spread over all machines, all very similar spec.
Mamajuanauk is the Name! Crunching is the Game!
  |
||
|
|
Sid2
Senior Cruncher USA Joined: Jun 12, 2007 Post Count: 259 Status: Offline Project Badges: 
|
CEP2 has just been announced as the WCG project in the BOINC Pentathlon and this isn't playing well in the stands:
----------------------------------------5/2/2013 7:36:11 AM | World Community Grid | No tasks are available for The Clean Energy Project - Phase 2  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well, was it a CPU or a GPU work-call? The system unfortunately will check all sciences selected even if there never was GPU work for a science, such as for CEP2. Just one line is not telling us... seldomly [not in my online dict] an isolated posted line tells us what's on.
----------------------------------------Whilst, there's not continuously everything in the feeder... a little patience is sometimes needed for slots to be vacated by assignment for other projects before new work can be backfilled. edit: seldom would suffice per various printed dicts as an adverb. [Edit 1 times, last edit by Former Member at May 2, 2013 1:07:19 PM] |
||
|
|
Sid2
Senior Cruncher USA Joined: Jun 12, 2007 Post Count: 259 Status: Offline Project Badges:
|
Thanks for the quick response!
----------------------------------------5/2/2013 8:07:22 AM | World Community Grid | update requested by user 5/2/2013 8:07:23 AM | World Community Grid | Sending scheduler request: Requested by user. 5/2/2013 8:07:23 AM | World Community Grid | Reporting 1 completed tasks, requesting new tasks for NVIDIA 5/2/2013 8:07:25 AM | World Community Grid | Scheduler request completed: got 0 new tasks 5/2/2013 8:07:25 AM | World Community Grid | No tasks sent 5/2/2013 8:07:25 AM | World Community Grid | No tasks are available for The Clean Energy Project - Phase 2 . . . apparently I'm just asking for GPU work. Unselected this preference: If my computer can process work on my graphics card, then please send me work to run on my graphics card for the projects that I have selected above. Now I get this: 5/2/2013 8:12:35 AM | World Community Grid | Requesting new tasks for NVIDIA 5/2/2013 8:12:37 AM | World Community Grid | Scheduler request completed: got 0 new tasks 5/2/2013 8:12:37 AM | World Community Grid | No tasks sent 5/2/2013 8:12:37 AM | World Community Grid | No tasks are available for The Clean Energy Project - Phase 2 5/2/2013 8:12:37 AM | World Community Grid | Tasks for NVIDIA GPU are available, but your preferences are set to not accept them 5/2/2013 8:12:37 AM | World Community Grid | Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them What am I doing wrong? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Nothing wrong! Your version of client does not tell why it's not fetching work for CPU. The new version 7, WCG kit expected soon, will say for instance "No work requested: Don't Need" which is stenography for "Buffer is already full". You can force the hand by first suspended work fetch for all the other attached projects to a client and then increase the minimum buffer a bit.
More refinement in server messages are needed. There is still "repair" work for HCC-GPU, but you don't want it, as per the prefs. |
||
|
|
Sid2
Senior Cruncher USA Joined: Jun 12, 2007 Post Count: 259 Status: Offline Project Badges:
|
Thanks, I'll have to muster up a little more patience and go fishing for HCC resends in the meantime.
----------------------------------------[Edit: 5/1/2013 9:46:37 PM | | Starting BOINC client version 7.0.28 for windows_x86_64 . . . is this version adequate?] [Edit 1 times, last edit by Sid2 at May 2, 2013 1:36:26 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
No, think David Anderson added [activated] these [about 12-14 different 'why' no work] in a later release and did an additional tweak for the 'Don't Need', which will be in a client *after* 7.0.64/65 general release and surely in the WCG version, since that one has not concluded yet.
As already elucidated, work fetch scheduling is separate for CPU and GPU, so if you have an idle GPU, the client will keep asking here or elsewhere [if allowed]. Soon as the CPU buffer side has space, it will fetch CEP2 work, noting that with use of the app_config.xml [as of 7.0.40], you can actually force a client to put CEP2 ahead, but that requires RTFMing and some skill development... it works very well with just 2 sciences, but can be done with 3 or more too, long as the total jobs always leave at least 1 CPU core free for CEP2 to take [use of <max_concurrent> in a deft way]. |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges: 
|
Unselected this preference: If my computer can process work on my graphics card, then please send me work to run on my graphics card for the projects that I have selected above. Now I get this: After the 1st. scheduler-request after the preference-change, the client should start using the new preference and stop asking for GPU-work from WCG... Even 7.0.28 is fairly old by now, AFAIK it should still stop asking for GPU-work.  "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges:
|
noting that with use of the app_config.xml [as of 7.0.40], you can actually force a client to put CEP2 ahead, Seeing he mentioned the Pentathlon, my guess is he'll run 100% CEP2, or CEP2 + NON-WCG, and in neither of these instances would using app_config.xml be much help... "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Feel free to *guess* and run CEP2 blindly on all cores, Ingleside, where WCG default continues to be one-at-the-time only! As it is, many systems have negative performance when running too many concurrent [my Linux box has]. Moreover, and most pronouncedly, the risk of more errors, heartbeat and too many exit crashes. So, back to my post, the recommended feature use with v7 to optimize and maximize CEP2 is with the utilization of app_config.xml for run of the mill systems in particular.
----------------------------------------[Edit 1 times, last edit by Former Member at May 2, 2013 3:57:50 PM] |
||
|
|
|