Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 9
|
![]() |
Author |
|
Scalextrix
Cruncher Joined: Oct 13, 2010 Post Count: 3 Status: Offline |
Hello i run tasks on my android devices. Im seeing issues with the deadline dates not downloading in date order.
For example yesterday i had tasks running with a deadline of july 22, i got new tasks today with deadline of july 15. The project should give equal time from download of the task otherwise i wont be able to finish the newer tasks in time. |
||
|
ca05065
Senior Cruncher Joined: Dec 4, 2007 Post Count: 325 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Assuming deadlines are the same for Android devices and PCs, the deadline for normal workunits for fight aids at home is 10 days and 4 days for resent workunits. A resend is issued if the original workunit has an error or is not returned by the deadline. Look in your results status section to see why.
|
||
|
Scalextrix
Cruncher Joined: Oct 13, 2010 Post Count: 3 Status: Offline |
Ok it appears that these are resent tasks, but giving less time to complete them means there is more chance they wont complete in time.
For PCs this would not seem to be so much of an issue but for android this does not make as much sense as the tasks take a lot longer to complete. I notice a lot of failed tasks on andriod, anything i can do about that? |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Resends have since a number of months a deadline which is 35 percent of the original. The agent knows the average daily computing time, so when there's a threat of not making it, the shorter deadline task will be prioritized.
----------------------------------------No, the android science app is pretty unstable. Nothing much you can do. Ditch wcg or reduce the number that run same time so the battery does 'not' deplete when plugged in and computing. Defenitely set only to compute when idle. This gives for me the highest success rate, if you're willing to put up with this resource wasting regimen. [Edit 1 times, last edit by Former Member at Jul 12, 2014 2:08:00 PM] |
||
|
Scalextrix
Cruncher Joined: Oct 13, 2010 Post Count: 3 Status: Offline |
I paused the 22nd July tasks so the 15 July tasks started processing, I have come back several hours later to find the longer deadline WUs have taken back the processor time. The system can prioritize on expected WU length and historic runtime, but as its a mobile device the system cant rely on how long it will be between processing sessions in the future. It would seem more sensible to always process shorter deadline tasks first.
I set the check to not compute when the screen is on, assuming thats what you mean by 'compute when idle'. I certainly found on a 2 core device that only processing on 1 core made a difference to reliability on several projects, I didnt notice the same multi-project issues on a 4 core device, but Ill experiment on the 4 core device now too. Thanks for the advice lavaflow. |
||
|
Eric_Kaiser
Veteran Cruncher Germany (Hessen) Joined: May 7, 2013 Post Count: 1047 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I wonder about the following:
----------------------------------------FAHV_x3I8WBprnotAS2_0561955_0027_2 was distributed to my newly attached android device HAL9000-S2 (added on August 24th) yesterday (August 25th). It was sent back this morning (8/26/2014 4:38:58 AM) and was marked as too late. ![]() |
||
|
Eric_Kaiser
Veteran Cruncher Germany (Hessen) Joined: May 7, 2013 Post Count: 1047 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Another one:
----------------------------------------Project Name: FightAIDS@Home - Vina Created: 08/24/2014 02:04:18 Name: FAHV_x3I8WAprnotAS2_0561423_0016 Minimum Quorum: 2 Replication: 3 Result Name App Version Number Status Sent Time Time Due / Return Time CPU Time / Elapsed Time (hours) Claimed/ Granted BOINC Credit FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 10-- 716 Error 26.08.14 18:01:12 26.08.14 19:37:47 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 9-- 716 Error 26.08.14 17:53:16 26.08.14 18:00:22 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 8-- 716 Error 26.08.14 17:38:57 26.08.14 17:52:37 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 7-- 716 Too Late 25.08.14 10:34:33 26.08.14 17:38:22 14.75 80.1 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 6-- 716 Error 25.08.14 02:51:48 25.08.14 10:33:21 6.58 31.9 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 5-- 716 Error 25.08.14 01:43:48 25.08.14 02:50:56 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 4-- 716 Error 25.08.14 01:40:26 25.08.14 01:42:37 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 3-- 716 Error 25.08.14 01:25:08 25.08.14 01:39:30 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 2-- 716 Error 25.08.14 01:21:51 25.08.14 01:24:35 0.00 0.0 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 1-- 716 Error 25.08.14 00:51:57 25.08.14 01:20:47 0.00 0.4 / 0.0 FAHV_ x3I8WAprnotAS2_ 0561423_ 0016_ 0-- 716 Too Late 25.08.14 00:51:04 26.08.14 07:26:29 12.98 53.9 / 0.0 Mine was the replication 7. ![]() |
||
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Too late could be from a number of things. What you are probably seeing is that a result was sent out to multiple hosts, but before your host could return i with success, all other copies up to the limit came back in either error, aborted, or detached. Thus the result was canceled on the sever side while yours was still working on the computation.
Thanks, -Uplinger |
||
|
Eric_Kaiser
Veteran Cruncher Germany (Hessen) Joined: May 7, 2013 Post Count: 1047 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Thank you very much for the update Uplinger.
----------------------------------------That proofs my conclusion. In the meanwhile a I've looked closer to these wu and I've seen that mostly my device was the only device that returned without error. ![]() |
||
|
|
![]() |