| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: Smash Childhood Cancer
Thread: SCC WU Length |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 25
|
|
| Author |
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1932 Status: Offline Project Badges: 
|
Don't know if you checked, but after the client was stopped, were there still processes running? No, I didn't check that. Those are things that IMHO should be resolved during the many Beta runs of the project...Ralf  |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
As you can see, when torsions and atoms get larger in the ligands the runtime increases. Unfortunately we can not size these. Oh, you might misunderstand me. That wasn't a complaint or a concern. I was just merely curious what the difference was with the increased lengths. :) Are the credits going to scale accordingly along with the increased times? Not that we all don't do these all out of the pure intent of heating up our homes somewhat during the winter. ;) That reminds me. Out of curiosity, I was looking at my results and noticed that sometimes I get less credit than claimed and sometimes more. Would explain that or point me to the appropriate post that explains this -/+ variation? ~Y Yavanius, Sorry about misunderstanding your question. If you are curious as to why the results are 2.5 times longer than the initial beta test. That is because the researchers have requested that the exhaustiveness parameter be set higher. This allows for a better accuracy in the results that are returned to them. In the first beta, the exhaustiveness was set to 8, they asked us to increase it to 20. With flexible dockings, higher exhaustiveness allows for more accurate results. As for credits, they will utilize the same formulas as the other vina applications, which should result in similar points per hour on results. For variation in the points, if you are compared against another user for your results, usually it is the average between the two that the results claimed. so if you claim 5 and the other user claims 7. Then both of you will be granted 6 as the work should be equal between the two. Thanks, -Uplinger |
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
Don't know if you checked, but after the client was stopped, were there still processes running? No, I didn't check that. Those are things that IMHO should be resolved during the many Beta runs of the project...Ralf Something to keep in mind is that stopping and restarting BOINC involves more than BOINCMgr. You can have BOINC crunching WUs on a machine without having BOINCMgr running. When you exit BOINCMgr, you normally get asked if you want to stop the rest of BOINC with it. You want to check that checkbox so that the BOINC process for each WU running is also stopped. |
||
|
|
Yavanius
Senior Cruncher Antarctica Joined: Jan 21, 2015 Post Count: 191 Status: Offline Project Badges: 
|
Sometimes WUs can still get stuck especially if BOINC core doesn't shut down properly or crashes, in which case you'll have to manually kill/end each app (if you don't want to restart).
|
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1932 Status: Offline Project Badges:
|
Something to keep in mind is that stopping and restarting BOINC involves more than BOINCMgr. You can have BOINC crunching WUs on a machine without having BOINCMgr running. When you exit BOINCMgr, you normally get asked if you want to stop the rest of BOINC with it. You want to check that checkbox so that the BOINC process for each WU running is also stopped. Well, as you could easily see with my badges, I am around here for a while (six years, one month and one week to be exact), just scraping along the Top 400 at WCG, as well as crunching for half a dozen other projects, though with lower priority as well. So I know very well the difference between closing the BOINCManager and closing the application(s) themselves...  Ralf  |
||
|
|
Yavanius
Senior Cruncher Antarctica Joined: Jan 21, 2015 Post Count: 191 Status: Offline Project Badges:
|
[]Well, as you could easily see with my badges, I am around here for a while (six years, one month and one week to be exact), just scraping along the Top 400 at WCG, ] So that's what that glare was... all those badges... Here, point them all over there, my house plants need some more sun... |
||
|
|
dcs1955
Veteran Cruncher USA Joined: May 24, 2016 Post Count: 668 Status: Offline Project Badges: 
|
This morning I went to check on desktops I have dedicated to WCG work. I usually just check to see that they have not be reset after an update. I saw that I had a SCC WU had been running for over 31 hrs. How long does it usually take for problems like this to get sorted out on new projects?
---------------------------------------- |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7581 Status: Offline Project Badges:
|
This morning I went to check on desktops I have dedicated to WCG work. I usually just check to see that they have not be reset after an update. I saw that I had a SCC WU had been running for over 31 hrs. How long does it usually take for problems like this to get sorted out on new projects? There is no perfect system to eliminate every error. I would suspect(without any proof) you had a work unit which somehow got hung in your machine. Generally speaking a reboot would fix this and the work unit would continue on normally. Others have reported this behavior and a reboot did fix it for them. Of the 4 million plus work units returned this appears to happen rarely. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Unless all SCC jobs display these ever extending runtimes [and lack of progress on the percentage, and lack of checkpointing], I'd refrain from booting 8-16-32-80 core devices, and first explore disabling LAIM (leave application in memory when suspended **), and then suspend the problem job(s) in question for 30 seconds [after first suspending all the 'ready to start' jobs], then restart them and see if crunching resumes, with the 'normal progress of precentage and checkpointing.
** New speak: Leave non-GPU task in memory when suspended. Old speak, how is BOINC installed... service/user, what security software could be taking a grip in the science app? Done 700 now on different OSses and never a blip/burp/belch. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Oh yes, the restarted jobs that properly left memory will display a drop back in processing time and possibly a little drop in progress %, worst case back to the beginning, but that always better than having all the jobs being restarted on a multithreaded device.
|
||
|
|
|