| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Drug Search for Leishmaniasis Forum
Thread: Looong running WUs |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 117
|
|
| Author |
|
|
RMau
Cruncher Joined: Feb 6, 2008 Post Count: 44 Status: Offline Project Badges: 
|
Here is an example of what I was seeing with DSFL on my AMD machines. Not all WUs were like this, but a majority of the ones I looked at were. And on all three machines.
----------------------------------------I'm open to explanations from the more knowledgeable techs and advisers for why this happens. And to Sek's point, yes the WUs did finish and were valid. It is just those seemingly lost hours between CPU time and Elapsed Time that bother me.  Rick  |
||
|
|
Mysteron347
Senior Cruncher Australia Joined: Apr 28, 2007 Post Count: 179 Status: Offline Project Badges: 
|
I'm open to explanations from the more knowledgeable techs and advisers for why this happens. And to Sek's point, yes the WUs did finish and were valid. It is just those seemingly lost hours between CPU time and Elapsed Time that bother me. I've just checked my machines (only 2 operating now - price of electricity...  )I see the same phenomenon. 2 cores with 55 mins CPU but 1h25 elapsed on a Core2Duo using XP, BOINC 6.10.58 target 14 These two tasks are running at High Priority and started immediately after being downloaded. My AMD machine is also running 4 tasks at High Priority (Win7/64, BOINC 6.12.26, all target 8) but all seems normal with these - a few seconds different between CPU and elapsed, but these units have been queued for a few days... About the only 'unusual' event I recall recently was an install of the latest Adobe reader. Such a minor event I'd normally dismiss it - I just know it happened at some time in the past 4 hours... Curious. |
||
|
|
deltavee
Ace Cruncher Texas Hill Country Joined: Nov 17, 2004 Post Count: 4894 Status: Offline Project Badges: 
|
The target 16 WUs are running considerable shorter. ~5-7 hours instead of 8-10 hours. Thanks uplinger.
|
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Yup, as of target 16, I have results finishing between 4 and 5 hours. Nice work techs and time to add it back to a couple slower machines.
---------------------------------------- Distributed computing volunteer since September 27, 2000 |
||
|
|
Gil II
Senior Cruncher Canada Joined: Dec 6, 2006 Post Count: 368 Status: Offline Project Badges: 
|
I have another long running WU.
----------------------------------------Results Status DSFL_ 00000017_ 0000031_ 0407_ 0-- Kermit In Progress 21/09/11 09:30:24 01/10/11 09:30:24 0.00 0.0 / 0.0 It has been running on my i7 for 11:18 hrs. and it is a 37.5%. My wingman's WU completed in 8 hrs. Shall I abort?  |
||
|
|
BSD
Senior Cruncher Joined: Apr 27, 2011 Post Count: 224 Status: Offline |
I would suspend the WU, reboot computer, resume WU.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I would........ first verify if the task has been checkpointing. First look in task properties when last and how much time has since gone by, then in the task slot output text file how often and last, activate the checkpoint writing to the message log. The how is described in the Start Here forum. That information will tell if the job is in loop di loop or just a hard crunch.
--//-- |
||
|
|
Gil II
Senior Cruncher Canada Joined: Dec 6, 2006 Post Count: 368 Status: Offline Project Badges:
|
This is really weird. Now it says that DSFL_ 00000017_ 0000031_ 0407_ 0 has been running for only 3:28 hrs and that it is at 46%. This sounds more credible but.... to go from saying that it has been running for 11+_ hours to only 3.5 hours?
----------------------------------------I am going to let it be... but still, this is weird. By the way, thanks SekerRob but I could not figure out how to look at the task properties. The Start here forum has way too many topics. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I'll frame that after screen-printing "Way too many topics in Start Here forum". If you open the index topic [stickied over all forums] and type "checkpoint", your browser will search it automatically... mine does.
In client 6.2 and up you select a running task, then hit the properties button on left. That's the way most anything is working in BOINC Manager at task/project level. --//-- |
||
|
|
Jason1478963
Senior Cruncher United States Joined: Sep 18, 2005 Post Count: 295 Status: Offline Project Badges: 
|
Some of this behavior reminds me of CEP2 where we are limiting the number of work units per machine to help keep the difference between cpu time and wall time down. Is this going to be required on this project as well?
---------------------------------------- |
||
|
|
|