| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: Beta test for Computing for Sustainable Water - Linux Only - May 7th 2012 - Issues Thread |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 39
|
|
| Author |
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges: 
|
!
----------------------------------------[Edit 6 times, last edit by skgiven at Jul 18, 2012 9:04:44 PM] |
||
|
|
BobCat13
Senior Cruncher Joined: Oct 29, 2005 Post Count: 295 Status: Offline Project Badges: 
|
Received 4 on a Q6600, suspended other tasks so all 4 would run together. All 4 completed without any problems.
----------------------------------------Interestingly, on 6.05 Beta and production, the runtimes varied from 3.47 to 3.63, but these 6.09 Betas have runtimes from 3.05 to 3.12. So are these sized smaller or is 6.09 more efficient? Edit: Finished 8 on the Q6600, running 4 at a time twice with no problems. All finished in the 3.05 to 3.12 time range. [Edit 1 times, last edit by BobCat13 at May 8, 2012 2:22:39 AM] |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Just got home and pushed them to the top on three separate machines. They'll all be running full bore simultaneously and I'll keep an eye on them.
----------------------------------------Not to beg for more results, but it seems to me that if the goal is to test full utilization there should have been many more units available to make sure they were taxing all the systems they reached. I have one machine that got zero because I wasn't here to trigger the work fetch and they all went relatively fast. Usually Linux only lasts much longer.  Distributed computing volunteer since September 27, 2000 |
||
|
|
evilkats
Senior Cruncher USA Joined: May 4, 2007 Post Count: 162 Status: Offline Project Badges: 
|
4 have just finished running side by side without any errors.
 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
All 8 of mine ran together without any issues. I hope this means I will soon be getting more than one CFSW at a time on my Linux boxes!
----------------------------------------[edit] Box is running Ubuntu 11.10 (Oneiric Ocelot) 64-bit, Intel i7 Quad-core with 16GB of memory.[/edit] [Edit 2 times, last edit by Former Member at May 8, 2012 3:04:52 AM] |
||
|
|
Dark Angel
Veteran Cruncher Australia Joined: Nov 11, 2005 Post Count: 721 Status: Offline Project Badges:
|
I had no issues that I saw running them on 32bit multicore and 64bit multicore Intel and AMD systems as well as on some virtual 64bit single and dual core machines.
---------------------------------------- Currently being moderated under false pretences |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Now, this is unfortunate. How do you end up with a daily quota of only 1 task?
07-May-2012 10:02:43 [World Community Grid] Computation for task BETA_cfsw_1000_01000268_0 finished 07-May-2012 10:02:43 [World Community Grid] Output file BETA_cfsw_1000_01000268_0_0 for task BETA_cfsw_1000_01000268_0 absent 07-May-2012 10:03:43 [World Community Grid] Sending scheduler request: To fetch work. Requesting 51840 seconds of work, reporting 1 completed tasks 07-May-2012 10:03:48 [World Community Grid] Scheduler request succeeded: got 0 new tasks 07-May-2012 10:03:48 [World Community Grid] Message from server: No tasks sent 07-May-2012 10:03:48 [World Community Grid] Message from server: No tasks are available for the applications you have selected. 07-May-2012 10:03:49 [World Community Grid] Message from server: This computer has finished a daily quota of 1 tasks |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Received 4 on a Q6600, suspended other tasks so all 4 would run together. All 4 completed without any problems. Interestingly, on 6.05 Beta and production, the runtimes varied from 3.47 to 3.63, but these 6.09 Betas have runtimes from 3.05 to 3.12. So are these sized smaller or is 6.09 more efficient? Edit: Finished 8 on the Q6600, running 4 at a time twice with no problems. All finished in the 3.05 to 3.12 time range. Sort of an educated guess looking at the batch number and where we are in progress (processed 0962-1044 in last 24 hours for CFSW), I'd say these were pulled straight from production, to be compared to results, which would also allow to check that nothing but the bug was changed and any optimizations did not effect the results.) Now with server 700, if there's promotion to production of 6.09, the system can match wingmen to use 6.05 for repairs i.e. would the results not be 100.0% compatible, there are not going to be invalids / inconclusives due a 6.05 - 6.09 meet. Processed 16 off the Octo, just got 2 more and 4 off the quad, and they did run without hickup and a fair bit faster... than the old Linux version, but don't have much material to compare on Linux since the trickle was just 1 per host for v 6.05. Certainly closer to Win, on the dual boot machine... Linux 64 - 11.10 BETA_ cfsw_ 1002_ 01002190_ 0-- 1767290 Pending Validation 7-5-12 23:18:20 8-5-12 04:43:30 4.90 99.8 / 0.0 BETA_ cfsw_ 1002_ 01002469_ 0-- 1767290 Pending Validation 7-5-12 23:18:20 8-5-12 04:38:25 4.80 98.1 / 0.0 BETA_ cfsw_ 1002_ 01002656_ 1-- 1767290 Pending Validation 7-5-12 23:18:20 8-5-12 04:36:06 4.75 97.5 / 0.0 BETA_ cfsw_ 1002_ 01002103_ 0-- 1767290 Pending Validation 7-5-12 23:18:20 8-5-12 04:30:17 4.70 96.0 / 0.0 W7-64 cfsw_ 1007_ 01007636_ 0-- 1112084 Valid 6-5-12 07:19:24 7-5-12 05:45:15 4.37 101.0 / 101.0 cfsw_ 1004_ 01004709_ 0-- 1112084 Valid 6-5-12 05:25:11 7-5-12 05:45:15 4.36 102.2 / 102.2 cfsw_ 1004_ 01004983_ 0-- 1112084 Valid 6-5-12 05:24:54 7-5-12 05:44:55 4.38 101.0 / 101.0 cfsw_ 1004_ 01004240_ 0-- 1112084 Valid 6-5-12 05:24:54 7-5-12 05:44:55 4.37 101.0 / 101.0 cfsw_ 1004_ 01004680_ 0-- 1112084 Valid 6-5-12 05:24:54 7-5-12 05:44:55 4.42 102.4 / 102.4 Think I'll stick to running this science on Windows, if techs deny these were differently sized. --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Now, this is unfortunate. How do you end up with a daily quota of only 1 task? 07-May-2012 10:02:43 [World Community Grid] Computation for task BETA_cfsw_1000_01000268_0 finished 07-May-2012 10:02:43 [World Community Grid] Output file BETA_cfsw_1000_01000268_0_0 for task BETA_cfsw_1000_01000268_0 absent 07-May-2012 10:03:43 [World Community Grid] Sending scheduler request: To fetch work. Requesting 51840 seconds of work, reporting 1 completed tasks 07-May-2012 10:03:48 [World Community Grid] Scheduler request succeeded: got 0 new tasks 07-May-2012 10:03:48 [World Community Grid] Message from server: No tasks sent 07-May-2012 10:03:48 [World Community Grid] Message from server: No tasks are available for the applications you have selected. 07-May-2012 10:03:49 [World Community Grid] Message from server: This computer has finished a daily quota of 1 tasks Jinxed, or the fact that your devices continue to proof they're hard to keep in production? Initial quota for a new science, to include version [which this beta is too] is 2 on the BOINC server 700 system. You failed 1, and used 1. My best guess --//-- edit: But then, my reliable rated Octo got max, twice, as did other testing members... but none reported error / invalid, which is the seeming key difference. [Edit 1 times, last edit by Former Member at May 8, 2012 6:44:39 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Just got home and pushed them to the top on three separate machines. They'll all be running full bore simultaneously and I'll keep an eye on them. Not to beg for more results, but it seems to me that if the goal is to test full utilization there should have been many more units available to make sure they were taxing all the systems they reached. I have one machine that got zero because I wasn't here to trigger the work fetch and they all went relatively fast. Usually Linux only lasts much longer. There were 3000 in quorum 2 put in the feeder (post uplinger). That's the approximate equivalent of a full Linux share production day, far bigger than any Linux share would ever have in a beta. These 6000 will give sufficient statistical information to see if this warrants the issue of a 6.1x beta, or moving 6.09 to production. And to follow the explanations on when 64 bits will be build needing passing thorugh Beta too, if not, it's a hint on how long this project is going to last (that is, if we would dream of seeing another 64 bit build science), but the begging is palpable. ;-) --//-- |
||
|
|
|