| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: Questions about participation drive for CEP2 |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 152
|
|
| Author |
|
|
martianmoons
Cruncher USA Joined: Nov 29, 2006 Post Count: 49 Status: Offline Project Badges: 
|
I use zero cache, which seems to work fine for me unless I am missing something. As soon as one work unit completes, another takes its place immediately for the most part.
|
||
|
|
anhhai
Veteran Cruncher Joined: Mar 22, 2005 Post Count: 839 Status: Offline Project Badges: 
|
caching is for when WCG goes down, like today where WCG went down for several hours. If I hadn't cache then my computer would of sit idle for hours. I believe that WCG advisers suggest keeping 1 days worth of cache, since it covers must cases of system failure
---------------------------------------- |
||
|
|
noderaser
Senior Cruncher United States Joined: Jun 6, 2006 Post Count: 297 Status: Offline Project Badges: 
|
You could always have a backup project, too... WCG is currently one of five projects I have running on my computers, but it ends up being mostly WCG since the others either send out work in short bursts, or are set as a lower priority than WCG--and thus only do work when BOINC tries to make up the credit debt or there is an outage at WCG.
-------------------------------------------------------------------------------- [Edit 2 times, last edit by noderaser at Jan 23, 2011 1:44:14 AM] |
||
|
|
martianmoons
Cruncher USA Joined: Nov 29, 2006 Post Count: 49 Status: Offline Project Badges:
|
Good point! Thanks.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You could always have a backup project, too... WCG is currently one of five projects I have running on my computers, but it ends up being mostly WCG since the others either send out work in short bursts, or are set as a lower priority than WCG--and thus only do work when BOINC tries to make up the credit debt or there is an outage at WCG. Nowadays "backup projects" can be really set to being exactly that. Only fetch work when WCG has run bone-dry. Set the Resource Share (Project Weight) on those to zero to achieve that. The 6.6 or higher versioned client will only crunch that work for as long as WCG is offline. Soon as WCG work is available again, the backup work will be paused and, I think, will be finished when the deadline comes near, or whatever you choose to do with those unfinished tasks. --//-- |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: 
|
I had series of WU's with the status "High Priority Running", and others that were normally running that were paused with the status "Waiting to Run". The reason as far as I could see was the return date limit of those WU's being very near to be reached.
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Repair jobs likely and your 3 day standard cache pushing it over into panic state. In a little while with 3 day cache you should not be receiving those short deadline jobs anymore. WCG tracks who consistently returning work < 2 days.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
How much resources does it take per CEP task? The "minimum requirements" are listed as 1 GB RAM and 2 GB disk, is this per task?
----------------------------------------So my i7 920 with 6 GB RAM should be able to run up to five of them simultaneously, leaving 1 GB for the OS and other non-Boinc tasks? [Edit 1 times, last edit by Former Member at Jan 23, 2011 2:02:01 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Those are the "controls" to get the jobs. WCG does not check if run multicore, so it's not checking like 4x1GB during work requests. That's BOINC jobs to see to it that enough is available when running and else you'll see what that other new thread is just having in it's title: "Waiting to run, Suspended, Waiting for Memory". In actuality, the jobs use substantially less with occasional spikes.
|
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges:
|
Dear bearcat, As skgiven wrote, you may see a substantial improvement if you only run a few CEP2 wus simultaneously and fill up your remaining cores with other projects which are less I/O intensive. Best wishes, Your Harvard CEP team Thanks for the reply. As I mentioned on my 1st post, on multicore systems (I have 4 core no HT, 8 core no HT and a 8 core w/HT -16 threads), when crunching just one along with other projects, have noticed the CPU time vs actual time grew more on the other tasks so it affects the other projects. I just started using my Mac mini, dual core only, and haven't test cep2 on this one but will see if I get the same kind of results as I do on my other pc's. Would be interesting to see if anyone with a SB chip has this kind of an issue. But it does seem the higher the core count, the larger the difference in CPU time vs actual time. Will let you know how the mini performs. Thanks again. Well, gave it a shot on the mini. 2.66 dual core. 25 minutes difference between CPU time vs actual time. Granted, only a 5400 rpm hard drive. Was crunching this and HFCC. Was hoping for a smaller difference but guess not. Might try later with a ssd drive I got cheap from micro center and see what happens.
Crunching for humanity since 2007!
 |
||
|
|
|