| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 8
|
|
| Author |
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges:
|
On ubuntu 64 11.4, seem to loose around 10 minutes between CPU time and actual time. This machine has a xeon x5650 hex. Currently using 11 threads, 10 on this project and 1 crunching climate prediction. Not sure if climate is the culprit or if it's this project issues. Anyone have this issue?
----------------------------------------
Crunching for humanity since 2007!
![]() |
||
|
|
RaymondFO
Veteran Cruncher USA Joined: Nov 30, 2004 Post Count: 561 Status: Offline Project Badges:
|
Mac mini had a huge time gap issue with GFAM and I reassigned the computer to another WU project, rebooted and all is well again.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Try reading this [in the GFAM forum]: http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,32061
----------------------------------------For terminology, actual time is called *elapsed* just as it's header says in BOINC Manager, the time the science was allowed to run. --//-- [Edit 1 times, last edit by Former Member at Nov 25, 2011 6:37:42 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Bearcat, would that "ubuntu 64 11.4" happen to be the same as, "Ubuntu 11.10 64-bit"?
; |
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges:
|
Ubuntu 11.10 is the next upgrade from 11.4.
----------------------------------------
Crunching for humanity since 2007!
![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
On an old P4 running WinXP SP3 I found that GFAM was about half as inefficient again as DSFL, which surprised me given that the techs have said that they were going to make a change to the code for both these projects so it's presumably identical.
Sorry I don't have the details any more; I already changed that machine to run DSFL only now. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Refer to my previous post in this thread give a link to another thread wherein one post says:
We are currently testing a fix for the write to disk issue in our internal Alpha testing environment for both DSFL and GOFAM. If things go well there should be a beta soon. Thanks, armstrdj That's AFAICD the only performance/efficiency improving fix worked on and discussed. Then, for those who continue to run with the 60 second *Write to Disk* default it won't make a difference anyhow if the fix is applied or not. Only for those that up it to a time that checkpoint writes are skipped... 5 minutes found to be a good number for those shutting down daily, but it can be set much larger for those running 24/7 and plan their boots. Anyhow, as eluded in that other thread, my Linux 64 bit 11.10 runs 99.2% efficient on DSFL, and a good percent lower for GFAM. Would expect for those with an SSD to easily match that number, given I'm on a > 5 year old Barracuda, albeit with optimized fstab settings in the OS [see other posts of what these are]. My even older W7-32 laptop does over 98% efficiency when left alone. --//-- |
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges:
|
Just started crunching this project on my win7 PC yesterday (22 threads). Looked at a completed wu and saw less than a minute difference. Currently crunching HCC and this project. Why such a difference between ubuntu and win7? Both 64.
----------------------------------------Am waiting on my ubuntu PC (11 threads) to finish crunching a climate prediction wu, then monitor to see if this has anything to do with it. Seems a bit weird to have this big of difference.
Crunching for humanity since 2007!
![]() |
||
|
|
|