| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: Mapping Cancer Markers Forum
Thread: workunit time to run |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 20
|
|
| Author |
|
|
mderrick
Cruncher Joined: May 8, 2007 Post Count: 7 Status: Offline Project Badges: 
|
Hi,
I've just unsubscribed from this project because work units are arriving with 7 days to run instead of the WCG default of 10 days. To keep these units means manual intervention. I'm happy to re-subscribe if someone can explain what I'm not getting right here ... Cheers, Michael |
||
|
|
supdood
Senior Cruncher USA Joined: Aug 6, 2015 Post Count: 333 Status: Offline Project Badges: 
|
Michael--The deadline for WUs is set for each individual project at WCG. Seven days is the deadline for MCM, so nothing wrong with what you are getting. If you are having trouble returning the WUs in seven days, you can reduce your queue to make sure that they get completed more quickly.
---------------------------------------- |
||
|
|
mderrick
Cruncher Joined: May 8, 2007 Post Count: 7 Status: Offline Project Badges:
|
Thanks for taking the time to reply, Supdood. I've been with WCG a while so I realise this. I was just intending to give the project owners feedback that shortening the deadline reduces the pool of users who can subscribe to a project. I already run at minimum queue. [I know this is quirky, but for environmental and machine longevity reasons I run Tthrottle to keep CPU temperature down to where natural cooling suffices - on my SP3, 55degC] The most successful BOINC project - Seti@home uses deadlines of 8 weeks or so! - they must have some reason for this ;)
|
||
|
|
mderrick
Cruncher Joined: May 8, 2007 Post Count: 7 Status: Offline Project Badges:
|
PS - my concerns aren't theoretical - I've cooked 2 laptops over the years!
|
||
|
|
James Lee72
Cruncher Joined: Jan 26, 2015 Post Count: 8 Status: Offline Project Badges:
|
It is truly unbelievable that anyone thinks that they should download the limit on WUs on this project. This project IS SO RELIABLE that my download reserve is set to .1 day. With 12 machines crunching, I worry more about Internet failure more than anything else. Even if I have a drive crash or a motherboard crash, there are only 4 - 8 WUs at risk. If this was Asteroids (or a few others) then, there would be a concern. Here, none. Set your reserves to a day or two if you are really concerned.
James |
||
|
|
Aurum
Master Cruncher The Great Basin Joined: Dec 24, 2017 Post Count: 2391 Status: Offline Project Badges: 
|
I find MCM is the best behaved of the WCG projects. It works great with queues as high as 2 days + 2 days. I now use 0.5 + 0.25 days in a futile attempt to get FAH2 to work, but I gave up as it's the worst WCG project.
----------------------------------------If you want a super reliable project try TN-GRID. They're studying transcription factors which is important fundamental life science research.   ...KRI please cancel all shadow-banning |
||
|
|
Luvfishin
Senior Cruncher Canada Joined: Apr 16, 2009 Post Count: 302 Status: Offline Project Badges: 
|
Where can I see the deadline of runtime for each individual project here? Thank you!
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I don't know of any page on this site which gives the current turn-round time for each project. That may be because the turn-round time for resends is (generally) lower and so adds another level of complexity to the information. Or it may be because it sometimes changes. But personally I think it would be very useful to have that information (both default and resend values) added as a table to the page at https://www.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq
I don't suppose the techs have time to do it, though. |
||
|
|
Luvfishin
Senior Cruncher Canada Joined: Apr 16, 2009 Post Count: 302 Status: Offline Project Badges:
|
I agree. I was saving working units for seven days, but I have now changed it to a maximum of two days after reading this thread. I didn't know that the units only lasted for a certain amount of time. The reason that I was setting for seven days was simply to maintain enough units to continue crunching if for some reason the project was having problems sending or if my Internet went down. However, it would still be nice to know the time limit. Thanks for your input. :-)
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I have my queues set to around 1.5 days, but that's probably rather long. The systems here are very reliable. There are outages for upgrades, but they are usually only a few hours. A bigger problem is when something unforeseen happens. There is automation to call the techs for something major, but the feeders can get clogged or run dry, and it can be the next day before anyone is around to kick them. And, as you say, internet connections can go down. But I reckon a day and shift covers pretty much everything, with the extra 4 hours to cover the possible change in run times from batch to batch. I can't remember a time when my systems ran dry set this way, but there were just a couple of times when it was getting close. But that's in many years of crunching here.
Hope that helps. |
||
|
|
|