| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Drug Search for Leishmaniasis Forum
Thread: Looong running WUs |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 117
|
|
| Author |
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
With people taking their machines off of DSFL, I hope they are letting the WUs finish or purging the ones they do not want to run. I do not want to see the WUs back up in PV until the time due is reached. The last three WUs I have in PV have CPU times of 10.13, 11.61, and 12.64. I still have target 001 WUs in PV jail. For me targets 1-4 ran much faster than anything else since. Even my fastest crunchers, which ran the early targets in 3.5-4 hours, are now taking 6.5-8.5 hours to get through target 11.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
Mysteron347
Senior Cruncher Australia Joined: Apr 28, 2007 Post Count: 179 Status: Offline Project Badges: 
|
Hmm. It seems this project isn't using the runtime-walls and generations model that HCMD2 was using - the length of this thread shows the reaction.
I would have expected that the methodology had been proved and returning to a suck-it-and-see method is a retrograde step. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It seems to me, Mysteron, that the techs can't win on this one.
On this sub forum, they're taking heat for not introducing a hard time stop on the units. If you visit HCMD-2 or CEP-2, you will find threads that frequently pop up to announce that the hard stops waste cycles and they want an option to run until its done, no matter how long it takes. This is the method the techs and scientists came up with for this project, and it was probably their best option. In short... suck it up |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
don't mind the long run's I presently have 2 pages finished setting in PV waiting on wingmen that also don't mind.
|
||
|
|
coolstream
Senior Cruncher SCOTLAND Joined: Nov 8, 2005 Post Count: 475 Status: Offline Project Badges: 
|
Thanks, Lawrence.
----------------------------------------Following up from your observation about HyperThreading, it might be useful for members to stae whether they are using it when discussing their runtimes. With HT on, I am finding that my runtimes are much higher than some of the reports I am reading in this thread. This doesn't bother me because each machine is still processing more data than if HT were switched off.  Crunching in memory of my Mum PEGGY, cousin ROPPA and Aunt AUDREY. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
coolstream
- You are correct and I agree 100%. turbo off they run fast on run longer but do more work |
||
|
|
RMau
Cruncher Joined: Feb 6, 2008 Post Count: 44 Status: Offline Project Badges: 
|
I stopped running DSFL. My machines that were running it are all AMD Quad Core. Two with Win 7, one with Ubuntu. As far as I know, Hyper Threading is an Intel thing. Unless there is a setting in BOINC or something?
----------------------------------------Rick  [Edit 1 times, last edit by RMau at Sep 14, 2011 12:34:42 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The fat lady sang her noon toon and it went "Ten hours twenty two minutes per score is a DSFL chore".
Yesterday just before booting into W7-64 cached a dozen target 12, batch 13 on Linux. Then in Windows ran a hand full of same 12-13 and finished in 11:30 hours average, range 10:41 - 11:59 hours. Now back to Linux to see how these do under Ubuntu 11.04, 64 bits. Better I think if only because the efficiency under Windows never hit 96% for the hands-off night run. Linux showed consistently > 99%. --//-- |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
The fat lady sang her noon toon and it went "Ten hours twenty two minutes per score is a DSFL chore". Yesterday just before booting into W7-64 cached a dozen target 12, batch 13 on Linux. Then in Windows ran a hand full of same 12-13 and finished in 11:30 hours average, range 10:41 - 11:59 hours. Now back to Linux to see how these do under Ubuntu 11.04, 64 bits. Better I think if only because the efficiency under Windows never hit 96% for the hands-off night run. Linux showed consistently > 99%. --//-- Sek, if your hands off Windows box only shows a 96% effeciency then there is something else running that's hoging resources. FWIW mine are all 99.4% or higher
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Very possibly the 3Tb shared drive indexing since it was 3 months since last time coming out of Linux AND the 1Gb of patches that Billy G decided to send down including SP1. As with Linux, I've just killed the power management on WIFI which certainly put the throughput close to the theoretical 300Mb max.[ot though]
--//-- |
||
|
|
|