| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: Computing for Clean Water 64 bit |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 81
|
|
| Author |
|
|
genes
Advanced Cruncher USA Joined: Jan 28, 2006 Post Count: 132 Status: Offline Project Badges: 
|
Remaining 2 finished and valid, 4.26 and 4.27 hours runtime.
---------------------- Win7 Ultimate x64 Intel Q6600, 4GB ram. BOINC mgr. 6.10.58 64-bit. |
||
|
|
gb009761
Master Cruncher Scotland Joined: Apr 6, 2005 Post Count: 2977 Status: Offline Project Badges: 
|
The repair WU I was fortunate to pick up, has now been duly returned
---------------------------------------- Result Name Device Name Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit BETA_ c4cw_ target03_ 061025069_ 1-- Valid 25/02/11 14:22:54 25/02/11 20:07:18 5.46 103.5 / 91.1 PS. Thanks for the b'day wishes   [Edit 1 times, last edit by gb009761 at Feb 25, 2011 10:41:08 PM] |
||
|
|
kateiacy
Veteran Cruncher USA Joined: Jan 23, 2010 Post Count: 1027 Status: Offline Project Badges:
|
Sorry to be late to the parade here, but it took me a couple days to obtain and run the last WUs that I needed in order to finish reporting comparisons on my two machines that got betas. I find the results very interesting, because the effect of going from 32- to 64-bit was more than 10 times greater on an AMD than an Intel processor. Running 32-bit code on a 64-bit AMD seems to really slow it down.
----------------------------------------AMD Phenom II X4 910e, 2.6 GHz: 14.4% speed-up BETA_ c4cw_ target03_ 061044260_ 1-- kate-amd64 Valid 2/25/11 02:00:23 2/25/11 21:55:11 4.26 109.2 / 91.1 BETA_ c4cw_ target03_ 061006324_ 0-- kate-amd64 Valid 2/24/11 03:33:45 2/24/11 12:31:18 4.28 108.4 / 89.6 BETA_ c4cw_ target03_ 061000474_ 0-- kate-amd64 Valid 2/24/11 03:34:05 2/24/11 12:31:18 4.22 107.0 / 89.7 c4cw_ target03_ 055474594_ 0-- kate-amd64 Pending Validation 2/24/11 14:46:04 2/25/11 04:32:56 4.97 125.8 / 0.0 Core 2 Duo P8700, 2.53 GHz 1.2% speed-up BETA_ c4cw_ target03_ 061005404_ 0-- system76-pc Valid 2/24/11 03:44:22 2/24/11 12:30:53 4.14 108.6 / 89.6 c4cw_ target03_ 056239288_ 0-- system76-pc Valid 2/25/11 12:34:07 2/26/11 21:57:08 4.19 109.9 / 90.6 c4cw_ target03_ 056223623_ 0-- system76-pc Valid 2/25/11 12:12:39 2/26/11 14:33:45 4.19 109.0 / 91.0 This was all run under Ubuntu 10.04 and BOINC 6.10.17. The only other post that I've seen in this thread with comparisons for an AMD Phenom showed a 15 or 16% speed-up.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
FYI: As of noon, about 6,450 results have validated for this run (90 this morning) at a mean run time of 4.43 hours. When this moves to production we'll be able to observe the true impact on Clean Water which still is running at a "flatter than Kansas" 5.0 hours per task.
Maybe WCG techs are willing to look and share what the Beta testing devices mean is for these hosts on the 6.15 production version to give a more accurate feel of the gain. --//-- PS, could be that the results over and above the 1000+5000 releases included the reissue of those first 150 that got linked to the 32 bit app. Then we're looking at a fail rate below the show stopper level... maybe already tomorrow we'll learn :p |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges: 
|
All Windows x64 systems:
Opteron (2.1GHz) Normal Tasks Valid 26/02/11 02:35:06 27/02/11 11:34:13 5.73 80.3 / 78.7 Valid 26/02/11 02:55:00 27/02/11 11:34:13 5.68 79.5 / 78.7 Valid 26/02/11 21:23:27 28/02/11 00:04:05 5.66 81.1 / 77.9 Avg. 5.69h, 80.3/78.4 Beta Tasks Valid 24/02/11 21:53:39 25/02/11 03:17:18 5.36 75.1 / 90.5 Valid 24/02/11 18:39:46 25/02/11 22:16:05 5.36 75.1 / 90.2 Avg. 5.36h, 75.1/90.4 6.1% faster, 15.3% more credit. Opteron (2.2GHz) Normal Tasks Valid 26/02/11 18:41:05 27/02/11 08:19:36 5.52 82.8 / 78.9 Valid 27/02/11 08:19:37 27/02/11 23:11:13 5.50 82.5 / 78.5 Valid 27/02/11 08:44:25 27/02/11 23:11:13 5.50 82.4 / 78.5 Avg. 5.51h, 82.6/78.6 Beta Tasks Valid 24/02/11 20:59:23 25/02/11 12:28:37 5.02 75.3 / 89.4 Avg. 5.02h, 75.3/89.4 9.8% faster, 13.7% more credit. Q8400 (2.66GHz) Normal Tasks Valid 27/02/11 18:38:24 27/02/11 22:40:06 3.95 84.9 / 79.5 Valid 27/02/11 04:29:12 27/02/11 09:04:20 3.85 82.6 / 79.0 Valid 27/02/11 00:27:15 27/02/11 04:28:54 3.79 81.4 / 79.4 Valid 27/02/11 00:25:36 27/02/11 04:22:37 3.78 81.2 / 79.1 Avg. 3.84h, 82.5/79.3 Beta Tasks Valid 24/02/11 21:52:43 25/02/11 08:35:54 3.79 81.3 / 90.2 Valid 24/02/11 21:48:08 25/02/11 07:27:32 3.79 81.4 / 90.1 Valid 24/02/11 21:48:08 25/02/11 04:50:12 3.71 79.7 / 89.5 Valid 24/02/11 20:52:18 25/02/11 00:43:59 3.80 81.6 / 90.7 Avg. 3.77h, 81.0/90.2 1.8% faster, 13.7% more credit. Phenom II 940 (3.0GHz + 4 GPU tasks) Normal Tasks Valid 27/02/11 09:50:27 27/02/11 15:40:10 4.26 85.0 / 80.7 Avg. 4.26, 85.0/80.7 Beta Tasks Valid 24/02/11 21:43:12 25/02/11 06:01:53 3.58 71.3 / 90.3 Valid 24/02/11 21:37:26 25/02/11 02:11:10 3.62 72.2 / 90.3 Valid 24/02/11 21:34:36 25/02/11 02:07:57 3.62 72.2 / 90.4 Avg. 3.61h, 71.9/90.3 18% faster, 11.9% more credit. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Don't attach any much meaning to the credit per task or hour during Beta as it's likely going to be recalibrated again when moved to production for this Zero Redundant science application. As we since recently know that the credit pools are per platform (Linux/Windows/Mac), it's kind of hoping that the 64 bits will get one designated too, which then means adding 4 more differentiations per science. Kind of complicates matters further with 8 sciences running on 3 or 4 platforms already (not all run on Mac PPC). We'll learn.
--//-- |
||
|
|
Jason1478963
Senior Cruncher United States Joined: Sep 18, 2005 Post Count: 295 Status: Offline Project Badges: 
|
Did anyone else experience errors on regular WCG work after the betas? I had 2 machines that usually run without errors. These 2 machines have experienced some errors that caused me to detach and attach to project again to solve the issue. These were Ubuntu machines that run at stock speeds. It could have been they started to panic if the internet wasn't working, but that usually doesn't require the detach & attach to fix the problem. I thought I would bring it up as it seems unusual for these machines to have errors like this.
---------------------------------------- |
||
|
|
Jason1478963
Senior Cruncher United States Joined: Sep 18, 2005 Post Count: 295 Status: Offline Project Badges:
|
I just had another one error all the work units. This one was a windows xp 64bit. Is anyone else seeing this? Two of them reporting application couldn't start. One windows XP 64bit the other Ubuntu 64 bit
----------------------------------------[Edit 1 times, last edit by Jason1478963 at Mar 1, 2011 7:21:08 PM] |
||
|
|
PMH_UK
Veteran Cruncher UK Joined: Apr 26, 2007 Post Count: 766 Status: Recently Active Project Badges: 
|
Believed to be unwanted effect of Key renewal, not seen in testing.
----------------------------------------Being discussed on DDDT2 forum: http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,30946 related to key renewal: http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,30943 Paul.
Paul.
|
||
|
|
Jason1478963
Senior Cruncher United States Joined: Sep 18, 2005 Post Count: 295 Status: Offline Project Badges:
|
Thank you for the help. This does seem to fit my issues
---------------------------------------- |
||
|
|
|