| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Human Proteome Folding - Phase 2
Thread: Recent HPF2 WU's tough for older equipment |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 6
|
|
| Author |
|
|
David_L6
Senior Cruncher USA Joined: Aug 24, 2006 Post Count: 296 Status: Offline Project Badges: 
|
Just an FYI. May want to make the HPF2 work units a little smaller for the older equipment out there.
----------------------------------------I have a couple of P4's that are taking a long time to chew through the recent HPF2 work units they received. Both P4's are running at 3.4GHz. For example: A P4 550 Prescott has been running two HPF2 work units for 18 hours and still lack 2+ and 3+ hours. The work units seems to be progressing normally, just slowly on these machines. My Core 2 Quads aren't having as difficult a time. The quads are getting the work units completed in ~ 5 to 7 hours depending upon the speed the individual machine is clocked at (from 2.4GHz to 3.6GHz). That is a long time compared to the other projects. HCC, DDDT, and FA@H all complete in less than 5 hours. Most complete in less than 4 hours. The difference is probably hyper-threading CPU's (one core that appears to be two) (P4's) versus four individual cores (Core 2 Quads). This post is not a gripe - just a heads up.  |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
P4HT, particular with heavies like HPF2 should be run single threaded.
----------------------------------------Because they are so hard to chew did WCG give them 20 days deadline. The DDDT/FAAH jobs have 14 days and HCC 10 days reflective of their shorter run times and to take account of the slower machines in the pack.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
David_L6
Senior Cruncher USA Joined: Aug 24, 2006 Post Count: 296 Status: Offline Project Badges:
|
Guess I jinxed myself! Found a hung WU this morning on one of my quads... 34 hours run time, 23% complete, 26 hours remaining. Shut down and re-started both BOINC and computer. Now it shows a little over 1 hour run time, 23.5% complete and ~ 1 hour remaining.
---------------------------------------- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello David_L6,
HPF2 does occasionally get caught in an infinite loop: http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=16378 Lately the Mac has been getting stuck on FAAH and DDDT. We hope that recompiling with the new BOINC API will cure that. If not, then we will look into it more after we roll out the new applications. Lawrence |
||
|
|
littlepeaks
Veteran Cruncher USA Joined: Apr 28, 2007 Post Count: 748 Status: Offline Project Badges: 
|
For what it's worth, I just got a HPF2 WU with an estimated completed time of almost 21 hours. That's a new record here!!! It is ls060_00047_9.
Running a Pentium 4 at 2.8 Ghz w/1.5 GB RAM. No problem here, except, will probably start backing off on the CPU time pretty soon when it starts getting warm (the fan starts really speeding up). |
||
|
|
littlepeaks
Veteran Cruncher USA Joined: Apr 28, 2007 Post Count: 748 Status: Offline Project Badges:
|
My PC completed the WU in 13.07 hours. So it must be that the original time calculation was not correct.
----------------------------------------[Edit 1 times, last edit by littlepeaks at Apr 29, 2008 3:03:31 AM] |
||
|
|
|