| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Human Proteome Folding - Phase 2
Thread: No 'finished' file... repeatedly |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 13
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well for one, you seem to have a setting that stops the agent when you use the computer: "Suspending computation - user is activ". For starters, would disable that feature as it likely unloads the job from memory. Secondly, suggest to activate the 'Leave in Memory". Even when work is paused, it will stay in ram, zero cpu time use, and be able to resume real quick. The settings in the device profile are: Do work while computer is in use? Yes Leave applications in memory while preempted? Yes give it a try and let us know: I'll give it a try but, that seems like it might interfere with my using my computer for possibly memory intensive things... Anyway, the problem started while I was out of town for four days. Will give it a try anyway. EDIT: Found another thread where someone had a similar looking problem. Based on what worked for him, I will try upping my swap when I get home. Not sure what it's set to at the moment. [Edit 1 times, last edit by Former Member at Apr 4, 2007 5:14:15 PM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
That's fine, but keep in mind that there are processes that BOINC could interpret at User Activity like some prominent screensavers.
----------------------------------------It's not coincidental that you get these "zero status but no 'finished' file" messages right after resume. Which are quite common, particular on machines with poor time keeping. As for your intensive tasks, BOINC goes incredibly quick into idle when the user works. Delays are mostly imperceptable. Your ram size and CPU power support the proposal. If still concerned, a 3rd party solution is Threadmaster. You can set that e.g. to 75% like the UD agent can (but only does on 1 core). All projects we have running are of a size that even if the heaviest is running on 2 cores simultaneous, does it not overly stretch your swap file. If your swapfile would run into limits, you will get quite more acute error/warning reporting. Would though be useful if you provide a link to the referenced thread. With 2gb ram and 9gb VM. That latter value (not knowing what your mativations were to set it that large), is normally severely over sized and more likely is source of system slow down. Suggest to visit the system control panel and set the minimum to RAM size, 2 gb and max to 4gb. Defrag prior and pagedefrag after (See Start Here, Useful Utilities thread). cheers
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
So far so good - it did, indeed, seem to demonstrate the problem when interrupted. I'll know for certain later today when everything finishes.
----------------------------------------Thanks! EDIT: That seems to have fixed it. I'm not home yet, but I see two results in "pending validation". Yay! :) [Edit 1 times, last edit by Former Member at Apr 5, 2007 8:51:00 PM] |
||
|
|
|