2 Clients Same WU different D/l times

Moderators: slegrand, Site Moderators, PandeGroup

2 Clients Same WU different D/l times

Postby patonb » Wed Mar 23, 2011 5:14 pm

Odd one today, both my gtx260s got the same WU, but at different times, 4min apart.

Never had this before, and both completed different units prior.

Gpu 1
Code: Select all
[16:43:32]   ... Done.
[16:43:32] DeleteFrameFiles: successfully deleted file=work/wudata_05.ckp
[16:43:32] Shutting down core
[16:43:32]
[16:43:32] Folding@home Core Shutdown: FINISHED_UNIT
[16:43:36] CoreStatus = 64 (100)
[16:43:36] Sending work to server
[16:43:36] Project: 10112 (Run 868, Clone 0, Gen 179)


[16:43:36] + Attempting to send results [March 23 16:43:36 UTC]
[16:43:38] + Results successfully sent
[16:43:38] Thank you for your contribution to Folding@Home.
[16:43:38] + Number of Units Completed: 549

[16:43:42] - Preparing to get new work unit...
[16:43:42] + Attempting to get work packet
[16:43:42] - Connecting to assignment server
[16:43:43] - Successful: assigned to (171.64.65.61).
[16:43:43] + News From Folding@Home: Welcome to Folding@Home
[16:43:43] Loaded queue successfully.
[16:43:44] + Closed connections
[16:43:44]
[16:43:44] + Processing work unit
[16:43:44] Core required: FahCore_11.exe
[16:43:44] Core found.
[16:43:44] Working on queue slot 06 [March 23 16:43:44 UTC]
[16:43:44] + Working ...
[16:43:44]
[16:43:44] *------------------------------*
[16:43:44] Folding@Home GPU Core
[16:43:44] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[16:43:44]
[16:43:44] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[16:43:44] Build host: amoeba
[16:43:44] Board Type: Nvidia
[16:43:44] Core      :
[16:43:44] Preparing to commence simulation
[16:43:44] - Looking at optimizations...
[16:43:44] DeleteFrameFiles: successfully deleted file=work/wudata_06.ckp
[16:43:44] - Created dyn
[16:43:44] - Files status OK
[16:43:44] - Expanded 73750 -> 383588 (decompressed 520.1 percent)
[16:43:44] Called DecompressByteArray: compressed_data_size=73750 data_size=383588, decompressed_data_size=383588 diff=0
[16:43:44] - Digital signature verified
[16:43:44]
[16:43:44] Project: 6601 (Run 2, Clone 353, Gen 776)
[16:43:44]
[16:43:44] Assembly optimizations on if available.
[16:43:44] Entering M.D.
[16:43:50] Tpr hash work/wudata_06.tpr:  2530165556 424065986 913645419 2189163447 4133208895
[16:43:50]
[16:43:50] Calling fah_main args: 14 usage=100
[16:43:50]
[16:43:51] Working on Protein
[16:43:52] Client config found, loading data.
[16:43:52] Starting GUI Server
[16:44:48] Completed 1%
[16:45:44] Completed 2%
[16:46:39] Completed 3%
[16:47:34] Completed 4%
[16:48:29] Completed 5%
[16:49:24] Completed 6%
[16:50:19] Completed 7%
[16:51:13] Completed 8%
[16:52:08] Completed 9%
[16:53:03] Completed 10%
[16:53:58] Completed 11%
[16:54:53] Completed 12%
[16:55:47] Completed 13%
[16:56:42] Completed 14%
[16:57:37] Completed 15%
[16:58:32] Completed 16%
[16:59:26] Completed 17%
[17:00:21] Completed 18%
[17:01:16] Completed 19%
[17:02:11] Completed 20%
[17:03:06] Completed 21%
[17:04:00] Completed 22%
[17:04:55] Completed 23%
[17:05:50] Completed 24%
[17:06:45] Completed 25%


Gpu 2

Code: Select all
[16:40:53]   ... Done.
[16:40:53] DeleteFrameFiles: successfully deleted file=work/wudata_06.ckp
[16:40:53] Shutting down core
[16:40:53]
[16:40:53] Folding@home Core Shutdown: FINISHED_UNIT
[16:40:56] CoreStatus = 64 (100)
[16:40:56] Unit 6 finished with 98 percent of time to deadline remaining.
[16:40:56] Updated performance fraction: 0.977338
[16:40:56] Sending work to server
[16:40:56] Project: 10112 (Run 520, Clone 1, Gen 114)


[16:40:56] + Attempting to send results [March 23 16:40:56 UTC]
[16:40:56] - Reading file work/wuresults_06.dat from core
[16:40:56]   (Read 123909 bytes from disk)
[16:40:56] Connecting to http://171.64.65.71:8080/
[16:40:58] Posted data.
[16:40:58] Initial: 0000; - Uploaded at ~61 kB/s
[16:40:58] - Averaged speed for that direction ~52 kB/s
[16:40:58] + Results successfully sent
[16:40:58] Thank you for your contribution to Folding@Home.
[16:40:58] + Number of Units Completed: 8800

[16:41:02] Trying to send all finished work units
[16:41:02] + No unsent completed units remaining.
[16:41:02] - Preparing to get new work unit...
[16:41:02] + Attempting to get work packet
[16:41:02] - Will indicate memory of 6141 MB
[16:41:02] - Connecting to assignment server
[16:41:02] Connecting to http://assign-GPU.stanford.edu:8080/
[16:41:03] Posted data.
[16:41:03] Initial: 40AB; - Successful: assigned to (171.64.65.61).
[16:41:03] + News From Folding@Home: Welcome to Folding@Home
[16:41:03] Loaded queue successfully.
[16:41:03] Connecting to http://171.64.65.61:8080/
[16:41:04] Posted data.
[16:41:04] Initial: 0000; - Receiving payload (expected size: 74262)
[16:41:05] - Downloaded at ~72 kB/s
[16:41:05] - Averaged speed for that direction ~84 kB/s
[16:41:05] + Received work.
[16:41:05] Trying to send all finished work units
[16:41:05] + No unsent completed units remaining.
[16:41:05] + Closed connections
[16:41:05]
[16:41:05] + Processing work unit
[16:41:05] Core required: FahCore_11.exe
[16:41:05] Core found.
[16:41:05] Working on queue slot 07 [March 23 16:41:05 UTC]
[16:41:05] + Working ...
[16:41:05] - Calling '.\FahCore_11.exe -dir work/ -suffix 07 -priority 96 -nocpulock -checkpoint 15 -verbose -lifeline 3764 -version 623'

[16:41:05]
[16:41:05] *------------------------------*
[16:41:05] Folding@Home GPU Core
[16:41:05] Version 1.31 (Tue Sep 15 10:57:42 PDT 2009)
[16:41:05]
[16:41:05] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
[16:41:05] Build host: amoeba
[16:41:05] Board Type: Nvidia
[16:41:05] Core      :
[16:41:05] Preparing to commence simulation
[16:41:05] - Looking at optimizations...
[16:41:05] DeleteFrameFiles: successfully deleted file=work/wudata_07.ckp
[16:41:05] - Created dyn
[16:41:05] - Files status OK
[16:41:05] - Expanded 73750 -> 383588 (decompressed 520.1 percent)
[16:41:05] Called DecompressByteArray: compressed_data_size=73750 data_size=383588, decompressed_data_size=383588 diff=0
[16:41:05] - Digital signature verified
[16:41:05]
[16:41:05] Project: 6601 (Run 2, Clone 353, Gen 776)
[16:41:05]
[16:41:05] Assembly optimizations on if available.
[16:41:05] Entering M.D.
[16:41:11] Tpr hash work/wudata_07.tpr:  2530165556 424065986 913645419 2189163447 4133208895
[16:41:11]
[16:41:11] Calling fah_main args: 14 usage=100
[16:41:11]
[16:41:11] Working on Protein
[16:41:12] Client config found, loading data.
[16:41:12] Starting GUI Server
[16:42:07] Completed 1%
[16:43:02] Completed 2%
[16:43:56] Completed 3%
[16:44:50] Completed 4%
[16:45:44] Completed 5%
[16:46:38] Completed 6%
[16:47:32] Completed 7%
[16:48:26] Completed 8%
[16:49:20] Completed 9%
[16:50:13] Completed 10%
[16:51:07] Completed 11%
[16:52:00] Completed 12%
[16:52:54] Completed 13%
[16:53:47] Completed 14%
[16:54:41] Completed 15%
[16:55:34] Completed 16%
[16:56:27] Completed 17%
[16:57:21] Completed 18%
[16:58:14] Completed 19%
[16:59:08] Completed 20%
[17:00:01] Completed 21%
[17:00:55] Completed 22%
[17:01:49] Completed 23%
[17:02:42] Completed 24%
[17:03:36] Completed 25%
[17:04:29] Completed 26%
[17:05:23] Completed 27%
[17:06:16] Completed 28%
WooHoo = L5639 @ 3.3Ghz Evga SR-2 6x2gb Corsair XMS3 CM 212+ Corsair 1050hx Blackhawk Ultra EVGA 560ti

Foldie = i7 950@ 4.0Ghz x58a-ud3r 216-216 @ 850/2000 3x2gb OCZ Gold NH-u12 Heatsink Corsair hx520 Antec 900
patonb
 
Posts: 1072
Joined: Thu Oct 23, 2008 2:42 am

Re: 2 Clients Same WU different D/l times

Postby i20234 » Wed Mar 23, 2011 5:20 pm

I don't see a problem with this. It probably is just that one gpu got a different WU than another previously which lead to the time change and/or one gpu was being used more than the other and/or your cpu gave one gpu higher priority/more resources to the other.
i20234
 
Posts: 56
Joined: Mon May 03, 2010 9:28 pm

Re: 2 Clients Same WU different D/l times

Postby topodisc » Wed Mar 23, 2011 5:36 pm

They have distinct machine id's I presume?
Image
Intel i7 2600K (D2) @ 4.60GHz ~ Noctua NH-D14 CPU cooler ~ Asus P8P67 Deluxe (B3/1305) ~ G.Skill 16GB (F3-12800CL7Q-16GBXH) ~ MSI GTX560 Ti @ 880MHz ~ Corsair TX650W PSU ~ Crucial C300 128GB SSD (0006) ~ Windows 7 Pro 64-bit
topodisc
 
Posts: 66
Joined: Sat Feb 12, 2011 3:38 am

Re: 2 Clients Same WU different D/l times

Postby patonb » Wed Mar 23, 2011 6:08 pm

i20234 wrote:I don't see a problem with this. It probably is just that one gpu got a different WU than another previously which lead to the time change and/or one gpu was being used more than the other and/or your cpu gave one gpu higher priority/more resources to the other.


You shouldn't usually have 2 wu the same unless you get extremely lucky, or the setup wrong.... I know why the times are different. That was brought up to show its not the exact same wu on the cards.i.e both cards grabed the 1 wu.

Yup, machine ids are different, have been folding on these for 3 months, and a year before on a differrent board. 16wu done on each since i rebooted yesterday... all different.

Its probably due to the closeness of the server call, it barfed... But Ii probably won't get pts for the 2nd one.

EDIT

And now its back to different units
patonb
 
Posts: 1072
Joined: Thu Oct 23, 2008 2:42 am

Re: 2 Clients Same WU different D/l times

Postby bruce » Wed Mar 23, 2011 7:14 pm

Multiple copies of the same WU circulating in FAH are not common, but they do happen when something happens to the initial assignment. Getting two copies assigned to hardware owned by the same person is even more unlikely, but I've seen a few reports of it happening. Fortunately, you DO get credit as long as the MachineIDs have been configured correctly.

WU assigned to donor at: 2011-03-23 09:41:13
Hi patonb (team 12864),
Your WU (P6601 R2 C353 G776) was added to the stats database on 2011-03-23 11:05:14 for 450 points of credit.

WU assigned to donor at: 2011-03-23 09:43:53
Hi patonb (team 12864),
Your WU (P6601 R2 C353 G776) was added to the stats database on 2011-03-23 11:05:14 for 450 points of credit.
bruce
 
Posts: 21332
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: 2 Clients Same WU different D/l times

Postby topodisc » Wed Mar 23, 2011 8:00 pm

Interesting that both copies are credited. Is issuing a WU to multiple machines used as a check that the WUs are being generated correctly (statistically for example)?
topodisc
 
Posts: 66
Joined: Sat Feb 12, 2011 3:38 am

Re: 2 Clients Same WU different D/l times

Postby bruce » Wed Mar 23, 2011 9:48 pm

If a WU is issued and completed, it is not reissued. If a WU is issued and it expires, it is reissued immediately. In some cases, it may be reissued more than once but I'm not sure about the differences between a simple reissue and a double reissue. Perhaps it depends on the amount of time it has been since the WU was issued the first time. Certainly if it has expired more than once, the need to assure a prompt completion increases to keep that trajectory from falling behind.
bruce
 
Posts: 21332
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: 2 Clients Same WU different D/l times

Postby patonb » Thu Mar 24, 2011 2:22 am

Yha it grabbed 2 different ones right after..

Thanks bruce.
patonb
 
Posts: 1072
Joined: Thu Oct 23, 2008 2:42 am


Return to NVIDIA specific issues

Who is online

Users browsing this forum: No registered users and 2 guests

cron