| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: GO Fight Against Malaria
Thread: Extremely slow / long running GFAM tasks |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 6
|
|
| Author |
|
|
MarshallW
Cruncher Joined: Nov 16, 2004 Post Count: 13 Status: Offline Project Badges: 
|
Looking at my current tasks on my main PC. I have 2 GFAM tasks out of 4 total tasks running. The first shows elapsed time of 126+ hours, with 332+ hours to go. It had a report deadline of 2 days ago and continues to run. Sort of. When I look at task properties, I see the same elapsed time, but the CPU time is only 27:04 out of over 126 hours of elapsed time, and it says it is only 6.379% complete.
The second task shows 73:48 elapsed, 41:12 to go, and CPU time is 4:37:11 out of those 73 hours of elapsed time, and that is only 62.963% done. On my laptop, I have another GFAM task showing very similar properties: 32 hours run, 53 to go, only 3:11:28 CPU time and only 27.6% done. What is going on here? Are they stuck? Should I cancel/abort these tasks? Why is one running 2 days after the report deadline? All other tasks from all other projects are running, completing, reporting, and being validated with now problems. But this has cut my throughput by 50%. Thanks for any help, MarshallW |
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1405 Status: Offline Project Badges: 
|
What is going on here? Are they stuck? Should I cancel/abort these tasks? It looks like they are stuck.Before aborting you could try to revive them by unticking in the preferences 'Leave applications in memory' and suspend the task. After 1 minute resume the task again and see, whether the task start consuming cpu cycles again. Why is one running 2 days after the report deadline? Because it started before the deadline. |
||
|
|
MarshallW
Cruncher Joined: Nov 16, 2004 Post Count: 13 Status: Offline Project Badges:
|
I will give it a try. Thanks for the advice.
|
||
|
|
MarshallW
Cruncher Joined: Nov 16, 2004 Post Count: 13 Status: Offline Project Badges:
|
I tried unchecking stay in memory, suspending, waiting a couple of minutes, and resuming. The GFAM tasks stayed stuck on both of my computers. I aborted them. Others have now started up, and are also stuck. Is anyone else seeing this behavior?
Is GFAM looking for some particular resource I might not have made available? If I can't fix it, I will abort them as before, and then uncheck GFAM from my list of projects. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
If you're not running BOINC CPU time at 100% (of idle spare cycles), then you may have walked into the known VINA engine based stuck problem seen with DSFL/GFAM and little/not heard of for SN2S. Remedy: Set CPU time to 100% and use alternate ways of reducing temps (which is the sole purpose of the CPU time limiter in BOINC). Windows crunchers are having good results by using TThrottle and setting a temp ceiling, e.g. 65-70C, to slow BOINC down, mostly needed on laptops.
----------------------------------------If setting 100% CPU time does not resolve the matter and needing the cooling function [TThrottle developed for BOINC not an option], then recommend deselecting GFAM or any of the other 2 mentioned above, until a fix is in place. SN2S seems to still run on low priority per the stats, so something is still being worked on for this VINA based science. Sorry. --//-- [Edit 2 times, last edit by Former Member at Apr 24, 2012 6:20:13 PM] |
||
|
|
oliverstirling
Advanced Cruncher United Kingdom Joined: May 7, 2007 Post Count: 107 Status: Offline Project Badges: 
|
I'm experiencing the same issues, both with GFAM and DSFL on one of my Windows systems. Thanks for the fix suggestion Sek, have upped the CPU time to 100% as per your suggestion and will see how that goes
----------------------------------------Update: so far no issues other than temp control as this error was occurring on a notebook. Resolved with a spare notebook cooler I had lying around [Edit 1 times, last edit by oliverstirling at May 1, 2012 10:12:23 AM] |
||
|
|
|