Page 1 of 1

[SOLVED] Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 3:05 pm
by Tim_H
I can't get this wu to scale past 2 cores.
system specs:
Dell Precision M6600
i7-2829QM (hyperthreading and turbo enabled)
16GB ram
windows 7

It was started using Arguments: -verbosity 9 -smp 6 -oneunit -verbosity 9. Then it was restarted using Arguments: -verbosity 9 -smp -verbosity 9 with no change.

I'm going to let it run, just wanted to report it.

Code: Select all

# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.34

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Users\twhicks\Downloads\smp
Executable: C:\Users\twhicks\Downloads\smp\FAH6.34-win32-SMP.exe
Arguments: -verbosity 9 -smp 6 -oneunit -verbosity 9 

[22:27:23] - Ask before connecting: No
[22:27:23] - User name: Tim_H (Team 37412)
[22:27:23] - User ID: 362917BC369EF771
[22:27:23] - Machine ID: 1
[22:27:23] 
[22:27:23] Loaded queue successfully.
[22:27:23] - Preparing to get new work unit...
[22:27:23] Cleaning up work directory
[22:27:23] - Autosending finished units... [January 9 22:27:23 UTC]
[22:27:23] Trying to send all finished work units
[22:27:23] + No unsent completed units remaining.
[22:27:23] - Autosend completed
[22:27:24] + Attempting to get work packet
[22:27:24] Passkey found
[22:27:24] - Will indicate memory of 16264 MB
[22:27:24] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 7
[22:27:24] - Connecting to assignment server
[22:27:24] Connecting to http://assign.stanford.edu:8080/
[22:27:24] Posted data.
[22:27:24] Initial: 8F80; - Successful: assigned to (128.143.199.97).
[22:27:24] + News From Folding@Home: Welcome to Folding@Home
[22:27:24] Loaded queue successfully.
[22:27:24] Sent data
[22:27:24] Connecting to http://128.143.199.97:8080/
[22:27:25] Posted data.
[22:27:25] Initial: 0000; - Receiving payload (expected size: 1766497)
[22:27:29] - Downloaded at ~431 kB/s
[22:27:29] - Averaged speed for that direction ~538 kB/s
[22:27:29] + Received work.
[22:27:29] + Closed connections
[22:27:29] 
[22:27:29] + Processing work unit
[22:27:29] Core required: FahCore_a3.exe
[22:27:29] Core found.
[22:27:29] Working on queue slot 08 [January 9 22:27:29 UTC]
[22:27:29] + Working ...
[22:27:29] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 6 -checkpoint 15 -verbose -lifeline 8476 -version 634'

[22:27:29] 
[22:27:29] *------------------------------*
[22:27:29] Folding@Home Gromacs SMP Core
[22:27:29] Version 2.27 (Dec. 15, 2010)
[22:27:29] 
[22:27:29] Preparing to commence simulation
[22:27:29] - Looking at optimizations...
[22:27:29] - Created dyn
[22:27:29] - Files status OK
[22:27:31] - Expanded 1765985 -> 2700832 (decompressed 152.9 percent)
[22:27:31] Called DecompressByteArray: compressed_data_size=1765985 data_size=2700832, decompressed_data_size=2700832 diff=0
[22:27:31] - Digital signature verified
[22:27:31] 
[22:27:31] Project: 7504 (Run 1, Clone 263, Gen 21)
[22:27:31] 
[22:27:31] Assembly optimizations on if available.
[22:27:31] Entering M.D.
[22:27:37] Mapping NT from 6 to 6 
[22:27:41] Completed 0 out of 500000 steps  (0%)
[22:46:36] Completed 5000 out of 500000 steps  (1%)
[23:04:21] Completed 10000 out of 500000 steps  (2%)
[23:22:12] Completed 15000 out of 500000 steps  (3%)
[23:39:54] Completed 20000 out of 500000 steps  (4%)
[23:57:30] Completed 25000 out of 500000 steps  (5%)
[00:15:18] Completed 30000 out of 500000 steps  (6%)
[00:33:08] Completed 35000 out of 500000 steps  (7%)
[00:50:47] Completed 40000 out of 500000 steps  (8%)
[01:08:30] Completed 45000 out of 500000 steps  (9%)
[01:26:24] Completed 50000 out of 500000 steps  (10%)
[01:44:09] Completed 55000 out of 500000 steps  (11%)
[02:01:54] Completed 60000 out of 500000 steps  (12%)
[02:19:44] Completed 65000 out of 500000 steps  (13%)
[02:37:22] Completed 70000 out of 500000 steps  (14%)
[02:55:03] Completed 75000 out of 500000 steps  (15%)
[03:12:54] Completed 80000 out of 500000 steps  (16%)
[03:30:46] Completed 85000 out of 500000 steps  (17%)
[03:48:35] Completed 90000 out of 500000 steps  (18%)
[04:06:14] Completed 95000 out of 500000 steps  (19%)
[04:24:06] Completed 100000 out of 500000 steps  (20%)
[04:27:24] - Autosending finished units... [January 10 04:27:24 UTC]
[04:27:24] Trying to send all finished work units
[04:27:24] + No unsent completed units remaining.
[04:27:24] - Autosend completed
[04:41:59] Completed 105000 out of 500000 steps  (21%)
[04:59:47] Completed 110000 out of 500000 steps  (22%)
[05:17:36] Completed 115000 out of 500000 steps  (23%)
[05:35:26] Completed 120000 out of 500000 steps  (24%)
[05:53:08] Completed 125000 out of 500000 steps  (25%)
[06:11:00] Completed 130000 out of 500000 steps  (26%)
[06:28:41] Completed 135000 out of 500000 steps  (27%)
[06:46:52] Completed 140000 out of 500000 steps  (28%)
[07:04:38] Completed 145000 out of 500000 steps  (29%)
[07:22:24] Completed 150000 out of 500000 steps  (30%)
[07:40:09] Completed 155000 out of 500000 steps  (31%)
[07:57:54] Completed 160000 out of 500000 steps  (32%)
[08:15:41] Completed 165000 out of 500000 steps  (33%)
[08:33:26] Completed 170000 out of 500000 steps  (34%)
[08:51:01] Completed 175000 out of 500000 steps  (35%)
[09:09:01] Completed 180000 out of 500000 steps  (36%)
[09:26:51] Completed 185000 out of 500000 steps  (37%)
[09:44:33] Completed 190000 out of 500000 steps  (38%)
[10:02:20] Completed 195000 out of 500000 steps  (39%)
[10:20:05] Completed 200000 out of 500000 steps  (40%)
[10:27:24] - Autosending finished units... [January 10 10:27:24 UTC]
[10:27:24] Trying to send all finished work units
[10:27:24] + No unsent completed units remaining.
[10:27:24] - Autosend completed
[10:37:49] Completed 205000 out of 500000 steps  (41%)
[10:55:26] Completed 210000 out of 500000 steps  (42%)
[11:13:06] Completed 215000 out of 500000 steps  (43%)
[11:30:50] Completed 220000 out of 500000 steps  (44%)
[11:48:33] Completed 225000 out of 500000 steps  (45%)
[12:06:17] Completed 230000 out of 500000 steps  (46%)
[12:23:59] Completed 235000 out of 500000 steps  (47%)
[12:41:41] Completed 240000 out of 500000 steps  (48%)
[12:59:24] Completed 245000 out of 500000 steps  (49%)
[13:17:09] Completed 250000 out of 500000 steps  (50%)
[13:34:52] Completed 255000 out of 500000 steps  (51%)
[13:52:34] Completed 260000 out of 500000 steps  (52%)
[14:11:19] Completed 265000 out of 500000 steps  (53%)
[14:12:33] Killing all core threads
[14:12:33] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown at user request.
[14:12:33] ***** Got a SIGTERM signal (2)
[14:12:33] Killing all core threads
[14:12:33] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown.


--- Opening Log file [January 10 14:12:38 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.34

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Users\twhicks\Downloads\smp
Executable: C:\Users\twhicks\Downloads\smp\FAH6.34-win32-SMP.exe
Arguments: -verbosity 9 -smp -verbosity 9 

[14:12:38] - Ask before connecting: No
[14:12:38] - User name: Tim_H (Team 37412)
[14:12:38] - User ID: 362917BC369EF771
[14:12:38] - Machine ID: 1
[14:12:38] 
[14:12:38] Loaded queue successfully.
[14:12:38] 
[14:12:38] - Autosending finished units... [January 10 14:12:38 UTC]
[14:12:38] Trying to send all finished work units
[14:12:38] + Processing work unit
[14:12:38] Core required: FahCore_a3.exe
[14:12:38] + No unsent completed units remaining.
[14:12:38] - Autosend completed
[14:12:38] Core found.
[14:12:38] Working on queue slot 08 [January 10 14:12:38 UTC]
[14:12:38] + Working ...
[14:12:38] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 8 -checkpoint 15 -verbose -lifeline 8304 -version 634'

[14:12:38] 
[14:12:38] *------------------------------*
[14:12:38] Folding@Home Gromacs SMP Core
[14:12:38] Version 2.27 (Dec. 15, 2010)
[14:12:38] 
[14:12:38] Preparing to commence simulation
[14:12:38] - Ensuring status. Please wait.
[14:12:48] - Looking at optimizations...
[14:12:48] - Working with standard loops on this execution.
[14:12:48] - Previous termination of core was improper.
[14:12:48] - Files status OK
[14:12:48] - Expanded 1765985 -> 2700832 (decompressed 152.9 percent)
[14:12:48] Called DecompressByteArray: compressed_data_size=1765985 data_size=2700832, decompressed_data_size=2700832 diff=0
[14:12:48] - Digital signature verified
[14:12:48] 
[14:12:48] Project: 7504 (Run 1, Clone 263, Gen 21)
[14:12:48] 
[14:12:48] Entering M.D.
[14:12:54] Using Gromacs checkpoints
[14:12:54] Mapping NT from 8 to 8 
[14:12:55] Resuming from checkpoint
[14:12:55] Verified work/wudata_08.log
[14:12:55] Verified work/wudata_08.trr
[14:12:55] Verified work/wudata_08.xtc
[14:12:55] Verified work/wudata_08.edr
[14:12:56] Completed 261420 out of 500000 steps  (52%)
[14:33:20] Completed 265000 out of 500000 steps  (53%)
*EDIT:*
Blonde moment:
I found the problem, somewhere along the lines I had set the affinity to only two cores.

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 3:21 pm
by PantherX
You mean to say that in the Task Manager, the CPU Usage is ~25% when folding this WU?

According to the FAHlog, this is how the FahCore_a3 is running:
6 CPUs -> [22:27:37] Mapping NT from 6 to 6
After restart
8 CPUs -> [14:12:54] Mapping NT from 8 to 8

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 3:34 pm
by ChelseaOilman
That's odd because everything looks normal in your log as far as I can tell. The log shows you went from using 6 cores to all 8 cores of your i7 CPU.
Arguments: -verbosity 9 -smp 6 -oneunit -verbosity 9
[22:27:29] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 6 -checkpoint 15 -verbose -lifeline 8476 -version 634'
[22:27:37] Mapping NT from 6 to 6

Arguments: -verbosity 9 -smp -verbosity 9
[14:12:38] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 8 -checkpoint 15 -verbose -lifeline 8304 -version 634'
[14:12:54] Mapping NT from 8 to 8
PantherX beat me to it.

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 4:07 pm
by Tim_H
PantherX wrote:You mean to say that in the Task Manager, the CPU Usage is ~25% when folding this WU?
Yes, that's exactly what I meant.

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 4:32 pm
by Mstenholm
The TPF supports the 25 % claim.

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 5:26 pm
by 7im
Server status shows a Min SMP setting of 2 (you need at least a dual core to get one). Maybe they screwed up and set a Max of 2 as well? Maybe an Mod/Admin type can bring this to the attention of the researcher?

And just for grins, please remove the double verbosity 9 flags. Thanks.

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 5:41 pm
by Tim_H
7im wrote:And just for grins, please remove the double verbosity 9 flags. Thanks.
didn't notice I had done that.
fixed

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 6:19 pm
by gwildperson
Tim_H wrote:
PantherX wrote:You mean to say that in the Task Manager, the CPU Usage is ~25% when folding this WU?
Yes, that's exactly what I meant.
According to Task Manager, what's happening to the other 75%? Is it Idle, or is some other process using it?

You may need to enable "show tasks from all users" to see the statistics for the idle pseudo-process.

Re: Project: 7504 (Run 1, Clone 263, Gen 21)

Posted: Tue Jan 10, 2012 6:39 pm
by Tim_H
Blonde moment:
I found the problem, somewhere along the lines I had set the affinity to only two cores.

working fine now.

thanks for the help everyone!