• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Folding Dragon

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Automata

Destroyer of Empires and Use
Joined
May 15, 2006
I've been trying to make one of these since reading this thread, but I'm having a bit of difficulty. I can't seem to fold it in the shape needed to get the most points, which is a bit frustrating.

folding_dragon.jpg


I got the server up and running back in December, but I stopped folding due to power cost issues shortly after. Since I had the install ready, I simply started the program and it continued happily along its way. I checked on it this morning to find that it had a whopping total of 481 points, which seems substantially lower than it should be. Here is the log from last night until right now:

Code:
# Linux SMP Console Edition ###################################################
###############################################################################

                       [email protected] Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: /mnt/hitachi/Folding
Executable: ./fah6
Arguments: -smp -verbosity 9 

[05:12:37] - Ask before connecting: No
[05:12:37] - User name: thideras (Team 32)
[05:12:37] - User ID: 1589BC1D4B3CD33F
[05:12:37] - Machine ID: 1
[05:12:37] 
[05:12:37] Loaded queue successfully.
[05:12:37] 
[05:12:37] + Processing work unit
[05:12:37] Core required: FahCore_a3.exe
[05:12:37] Core found.
[05:12:37] - Autosending finished units... [February 15 05:12:37 UTC]
[05:12:37] Trying to send all finished work units
[05:12:37] + No unsent completed units remaining.
[05:12:37] - Autosend completed
[05:12:37] Working on queue slot 01 [February 15 05:12:37 UTC]
[05:12:37] + Working ...
[05:12:37] - Calling './FahCore_a3.exe -dir work/ -nice 19 -suffix 01 -np 4 -nocpulock -checkpoint 15 -verbose -lifeline 17677 -version 629'

Warning: Ignoring unknown arg 
[05:12:38] 
[05:12:38] *------------------------------*
[05:12:38] [email protected] Gromacs SMP Core
[05:12:38] Version 2.22 (June 10, 2010)
[05:12:38] 
[05:12:38] Preparing to commence simulation
[05:12:38] - Ensuring status. Please wait.
[05:12:47] - Looking at optimizations...
[05:12:47] - Working with standard loops on this execution.
[05:12:47] - Previous termination of core was improper.
[05:12:47] - Files status OK
[05:12:48] - Expanded 1761928 -> 2246261 (decompressed 127.4 percent)
[05:12:48] Called DecompressByteArray: compressed_data_size=1761928 data_size=2246261, decompressed_data_size=2246261 diff=0
[05:12:48] - Digital signature verified
[05:12:48] 
[05:12:48] Project: 6071 (Run 0, Clone 73, Gen 332)
[05:12:48] 
[05:12:48] Entering M.D.
Starting 4 threads
NNODES=4, MYRANK=0, HOSTNAME=thread #0
Reading file work/wudata_01.tpr, VERSION 4.0.99_development_20090605 (single precision)
NNODES=4, MYRANK=1, HOSTNAME=thread #1
NNODES=4, MYRANK=2, HOSTNAME=thread #2
NNODES=4, MYRANK=3, HOSTNAME=thread #3

Reading checkpoint file work/wudata_01.cpt generated: Mon Feb 14 23:01:10 2011

[05:12:54] Using Gromacs checkpoints
Making 1D domain decomposition 1 x 4 x 1
starting mdrun 'Mutant_scan'
166500016 steps, 333000.0 ps (continuing from step 166002087, 332004.2 ps).
[05:12:54] Resuming from checkpoint
[05:12:54] Verified work/wudata_01.log
[05:12:54] Verified work/wudata_01.trr
[05:12:54] Verified work/wudata_01.edr
[05:12:55] Completed 2071 out of 500000 steps  (0%)
[05:17:50] Completed 5000 out of 500000 steps  (1%)
[05:26:17] Completed 10000 out of 500000 steps  (2%)
[05:34:49] Completed 15000 out of 500000 steps  (3%)
[05:43:22] Completed 20000 out of 500000 steps  (4%)
[05:51:52] Completed 25000 out of 500000 steps  (5%)
[06:00:17] Completed 30000 out of 500000 steps  (6%)
[06:08:51] Completed 35000 out of 500000 steps  (7%)
[06:17:20] Completed 40000 out of 500000 steps  (8%)
[06:25:48] Completed 45000 out of 500000 steps  (9%)
[06:34:16] Completed 50000 out of 500000 steps  (10%)
[06:42:50] Completed 55000 out of 500000 steps  (11%)
[06:51:19] Completed 60000 out of 500000 steps  (12%)
[06:59:47] Completed 65000 out of 500000 steps  (13%)
[07:08:19] Completed 70000 out of 500000 steps  (14%)
[07:16:45] Completed 75000 out of 500000 steps  (15%)
[07:25:17] Completed 80000 out of 500000 steps  (16%)
[07:33:48] Completed 85000 out of 500000 steps  (17%)
[07:42:16] Completed 90000 out of 500000 steps  (18%)
[07:50:45] Completed 95000 out of 500000 steps  (19%)
[07:59:10] Completed 100000 out of 500000 steps  (20%)
[08:07:40] Completed 105000 out of 500000 steps  (21%)
[08:16:10] Completed 110000 out of 500000 steps  (22%)
[08:24:38] Completed 115000 out of 500000 steps  (23%)
[08:33:08] Completed 120000 out of 500000 steps  (24%)
[08:41:37] Completed 125000 out of 500000 steps  (25%)
[08:50:02] Completed 130000 out of 500000 steps  (26%)
[08:58:26] Completed 135000 out of 500000 steps  (27%)
[09:06:59] Completed 140000 out of 500000 steps  (28%)
[09:15:27] Completed 145000 out of 500000 steps  (29%)
[09:23:53] Completed 150000 out of 500000 steps  (30%)
[09:32:16] Completed 155000 out of 500000 steps  (31%)
[09:40:40] Completed 160000 out of 500000 steps  (32%)
[09:49:06] Completed 165000 out of 500000 steps  (33%)
[09:57:31] Completed 170000 out of 500000 steps  (34%)
[10:06:01] Completed 175000 out of 500000 steps  (35%)
[10:14:26] Completed 180000 out of 500000 steps  (36%)
[10:22:46] Completed 185000 out of 500000 steps  (37%)
[10:31:06] Completed 190000 out of 500000 steps  (38%)
[10:39:24] Completed 195000 out of 500000 steps  (39%)
[10:47:42] Completed 200000 out of 500000 steps  (40%)
[10:56:05] Completed 205000 out of 500000 steps  (41%)
[11:04:33] Completed 210000 out of 500000 steps  (42%)
[11:12:37] - Autosending finished units... [February 15 11:12:37 UTC]
[11:12:37] Trying to send all finished work units
[11:12:37] + No unsent completed units remaining.
[11:12:37] - Autosend completed
[11:13:09] Completed 215000 out of 500000 steps  (43%)
[11:21:34] Completed 220000 out of 500000 steps  (44%)
[11:29:54] Completed 225000 out of 500000 steps  (45%)
[11:38:18] Completed 230000 out of 500000 steps  (46%)
[11:46:42] Completed 235000 out of 500000 steps  (47%)
[11:55:05] Completed 240000 out of 500000 steps  (48%)
[12:03:32] Completed 245000 out of 500000 steps  (49%)
[12:11:59] Completed 250000 out of 500000 steps  (50%)
[12:20:24] Completed 255000 out of 500000 steps  (51%)
[12:28:46] Completed 260000 out of 500000 steps  (52%)
[12:37:07] Completed 265000 out of 500000 steps  (53%)
[12:45:32] Completed 270000 out of 500000 steps  (54%)
[12:54:03] Completed 275000 out of 500000 steps  (55%)
[13:02:35] Completed 280000 out of 500000 steps  (56%)
[13:11:07] Completed 285000 out of 500000 steps  (57%)
[13:19:34] Completed 290000 out of 500000 steps  (58%)
[13:28:04] Completed 295000 out of 500000 steps  (59%)
[13:36:30] Completed 300000 out of 500000 steps  (60%)
[13:45:01] Completed 305000 out of 500000 steps  (61%)
[13:53:28] Completed 310000 out of 500000 steps  (62%)
[14:02:00] Completed 315000 out of 500000 steps  (63%)
[14:10:29] Completed 320000 out of 500000 steps  (64%)
[14:19:04] Completed 325000 out of 500000 steps  (65%)
[14:27:33] Completed 330000 out of 500000 steps  (66%)
[14:36:02] Completed 335000 out of 500000 steps  (67%)
[14:44:29] Completed 340000 out of 500000 steps  (68%)
[14:52:59] Completed 345000 out of 500000 steps  (69%)
[15:01:29] Completed 350000 out of 500000 steps  (70%)
[15:09:59] Completed 355000 out of 500000 steps  (71%)
[15:18:26] Completed 360000 out of 500000 steps  (72%)
[15:26:52] Completed 365000 out of 500000 steps  (73%)
[15:35:20] Completed 370000 out of 500000 steps  (74%)
[15:43:49] Completed 375000 out of 500000 steps  (75%)
[15:52:17] Completed 380000 out of 500000 steps  (76%)
[16:00:50] Completed 385000 out of 500000 steps  (77%)
[16:09:17] Completed 390000 out of 500000 steps  (78%)
[16:17:49] Completed 395000 out of 500000 steps  (79%)
[16:26:17] Completed 400000 out of 500000 steps  (80%)
[16:34:46] Completed 405000 out of 500000 steps  (81%)
[16:43:16] Completed 410000 out of 500000 steps  (82%)
[16:51:47] Completed 415000 out of 500000 steps  (83%)
[17:00:15] Completed 420000 out of 500000 steps  (84%)
[17:08:44] Completed 425000 out of 500000 steps  (85%)
[17:12:37] - Autosending finished units... [February 15 17:12:37 UTC]
[17:12:37] Trying to send all finished work units
[17:12:37] + No unsent completed units remaining.
[17:12:37] - Autosend completed
[17:17:10] Completed 430000 out of 500000 steps  (86%)
[17:25:40] Completed 435000 out of 500000 steps  (87%)
[17:34:13] Completed 440000 out of 500000 steps  (88%)
[17:42:37] Completed 445000 out of 500000 steps  (89%)
[17:51:04] Completed 450000 out of 500000 steps  (90%)
[17:59:32] Completed 455000 out of 500000 steps  (91%)
[18:08:04] Completed 460000 out of 500000 steps  (92%)
[18:16:38] Completed 465000 out of 500000 steps  (93%)
[18:25:10] Completed 470000 out of 500000 steps  (94%)
[18:33:42] Completed 475000 out of 500000 steps  (95%)
[18:42:14] Completed 480000 out of 500000 steps  (96%)
[18:50:46] Completed 485000 out of 500000 steps  (97%)
[18:59:17] Completed 490000 out of 500000 steps  (98%)
[19:07:50] Completed 495000 out of 500000 steps  (99%)
[19:16:24] Completed 500000 out of 500000 steps  (100%)

Writing final coordinates.

 Average load imbalance: 5.3 %
 Part of the total run time spent waiting due to load imbalance: 2.0 %


    Parallel run - timing based on wallclock.

               NODE (s)   Real (s)      (%)
       Time:  50610.565  50610.565    100.0
                       14h03:30
               (Mnbf/s)   (GFlops)   (ns/day)  (hour/ns)
Performance:    119.830      7.744      1.700     14.117

Thanx for Using GROMACS - Have a Nice Day

[19:16:25] DynamicWrapper: Finished Work Unit: sleep=10000
[19:16:35] 
[19:16:35] Finished Work Unit:
[19:16:35] - Reading up to 3697488 from "work/wudata_01.trr": Read 3697488
[19:16:35] trr file hash check passed.
[19:16:35] edr file hash check passed.
[19:16:35] logfile size: 58390
[19:16:35] Leaving Run
[19:16:36] - Writing 3791430 bytes of core data to disk...
[19:16:37]   ... Done.
[19:17:15] - Shutting down core
[19:17:15] 
[19:17:15] [email protected] Core Shutdown: FINISHED_UNIT
[19:17:20] CoreStatus = 64 (100)
[19:17:20] Unit 1 finished with 90 percent of time to deadline remaining.
[19:17:20] Updated performance fraction: 0.862486
[19:17:20] Sending work to server
[19:17:20] Project: 6071 (Run 0, Clone 73, Gen 332)


[19:17:20] + Attempting to send results [February 15 19:17:20 UTC]
[19:17:20] - Reading file work/wuresults_01.dat from core
[19:17:20]   (Read 3791430 bytes from disk)
[19:17:20] Connecting to http://171.64.65.54:8080/
[19:17:50] Posted data.
[19:17:51] Initial: 0000; - Uploaded at ~119 kB/s
[19:17:51] - Averaged speed for that direction ~278 kB/s
[19:17:51] + Results successfully sent
[19:17:51] Thank you for your contribution to [email protected]
[19:17:51] + Number of Units Completed: 20

Warning: Ignoring unknown arg 
[19:18:02] Trying to send all finished work units
[19:18:02] + No unsent completed units remaining.
[19:18:02] - Preparing to get new work unit...
[19:18:02] Cleaning up work directory
[19:18:04] + Attempting to get work packet
[19:18:04] - Will indicate memory of 7983 MB
[19:18:04] - Connecting to assignment server
[19:18:04] Connecting to http://assign.stanford.edu:8080/
[19:18:04] Posted data.
[19:18:04] Initial: 40AB; - Successful: assigned to (171.64.65.54).
[19:18:04] + News From [email protected]: Welcome to [email protected]
[19:18:05] Loaded queue successfully.
[19:18:05] Connecting to http://171.64.65.54:8080/
[19:18:05] Posted data.
[19:18:05] Initial: 0000; - Receiving payload (expected size: 1765267)
[19:18:07] - Downloaded at ~861 kB/s
[19:18:07] - Averaged speed for that direction ~1028 kB/s
[19:18:07] + Received work.
[19:18:07] Trying to send all finished work units
[19:18:07] + No unsent completed units remaining.
[19:18:07] + Closed connections
[19:18:07] 
[19:18:07] + Processing work unit
[19:18:07] Core required: FahCore_a3.exe
[19:18:07] Core found.
[19:18:07] Working on queue slot 02 [February 15 19:18:07 UTC]
[19:18:07] + Working ...
[19:18:07] - Calling './FahCore_a3.exe -dir work/ -nice 19 -suffix 02 -np 4 -nocpulock -checkpoint 15 -verbose -lifeline 17677 -version 629'

Warning: Ignoring unknown arg 
[19:18:07] 
[19:18:07] *------------------------------*
[19:18:07] [email protected] Gromacs SMP Core
[19:18:07] Version 2.22 (June 10, 2010)
[19:18:07] 
[19:18:07] Preparing to commence simulation
[19:18:07] - Looking at optimizations...
[19:18:07] - Created dyn
[19:18:07] - Files status OK
[19:18:08] - Expanded 1764755 -> 2251189 (decompressed 127.5 percent)
[19:18:08] Called DecompressByteArray: compressed_data_size=1764755 data_size=2251189, decompressed_data_size=2251189 diff=0
[19:18:08] - Digital signature verified
[19:18:08] 
[19:18:08] Project: 6065 (Run 0, Clone 113, Gen 133)
[19:18:08] 
[19:18:08] Assembly optimizations on if available.
[19:18:08] Entering M.D.
Starting 4 threads
NNODES=4, MYRANK=1, HOSTNAME=thread #1
NNODES=4, MYRANK=0, HOSTNAME=thread #0
Reading file work/wudata_02.tpr, VERSION 4.0.99_development_20090605 (single precision)
NNODES=4, MYRANK=3, HOSTNAME=thread #3
NNODES=4, MYRANK=2, HOSTNAME=thread #2
Making 1D domain decomposition 1 x 4 x 1
starting mdrun 'Mutant_scan'
67000008 steps, 134000.0 ps (continuing from step 66500008, 133000.0 ps).
[19:18:14] Completed 0 out of 500000 steps  (0%)
[19:26:41] Completed 5000 out of 500000 steps  (1%)
[19:35:14] Completed 10000 out of 500000 steps  (2%)
[19:43:43] Completed 15000 out of 500000 steps  (3%)
[19:52:13] Completed 20000 out of 500000 steps  (4%)
[20:00:50] Completed 25000 out of 500000 steps  (5%)
[20:09:23] Completed 30000 out of 500000 steps  (6%)
[20:17:53] Completed 35000 out of 500000 steps  (7%)
[20:26:25] Completed 40000 out of 500000 steps  (8%)
[20:34:57] Completed 45000 out of 500000 steps  (9%)
[20:43:25] Completed 50000 out of 500000 steps  (10%)
[20:51:55] Completed 55000 out of 500000 steps  (11%)
[21:00:24] Completed 60000 out of 500000 steps  (12%)
[21:09:01] Completed 65000 out of 500000 steps  (13%)
[21:17:36] Completed 70000 out of 500000 steps  (14%)
[21:26:04] Completed 75000 out of 500000 steps  (15%)
[21:34:34] Completed 80000 out of 500000 steps  (16%)
[21:43:03] Completed 85000 out of 500000 steps  (17%)
[21:51:35] Completed 90000 out of 500000 steps  (18%)
[22:00:07] Completed 95000 out of 500000 steps  (19%)
[22:08:36] Completed 100000 out of 500000 steps  (20%)
[22:17:08] Completed 105000 out of 500000 steps  (21%)
[22:25:43] Completed 110000 out of 500000 steps  (22%)
[22:34:13] Completed 115000 out of 500000 steps  (23%)
[22:42:42] Completed 120000 out of 500000 steps  (24%)
[22:51:12] Completed 125000 out of 500000 steps  (25%)
[22:59:33] Completed 130000 out of 500000 steps  (26%)
[23:06:33] Completed 135000 out of 500000 steps  (27%)
[23:12:37] - Autosending finished units... [February 15 23:12:37 UTC]
[23:12:37] Trying to send all finished work units
[23:12:37] + No unsent completed units remaining.
[23:12:37] - Autosend completed
[23:13:26] Completed 140000 out of 500000 steps  (28%)
[23:20:20] Completed 145000 out of 500000 steps  (29%)
Both work units are reporting 481 points, per this website. Am I missing something or did they really lower the value of running SMP on a quad core?

The server is based around an AMD Phenom x4 2.6GHz BE. The system has 8gb of RAM and is naturally running a healthy amount of hard drives due to its job as a file server.

Do I need to use the Passkey to get extra point on a non-bigadv computer? I didn't think this was needed.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

My main system, on the other hand, is fighting with me tooth and claw. I can't even get the CPU or GPU client to do anything. My main system is running a QX9650 at 4.0GHz with 8gb of RAM and a GTX 260. The operating system of choice is Fedora 14 running the 2.6.35.10-74.fc14.x86_64 kernel. I found a few posts that had some issues launching the client and found that I needed to downgrade to the 6.02 version, which actually fixed my first problem. Now, the client can't connect to the server to pull down any work.

Code:
[[email protected] CPU]$ ./fah6 -smp -verbosity 9

Note: Please read the license agreement (fah6 -license). Further 
use of this software requires that you have read and accepted this agreement.

4 cores detected


--- Opening Log file [February 15 23:20:49] 


# SMP Client ##################################################################
###############################################################################

                       [email protected] Client Version 6.02

                          http://folding.stanford.edu

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

Launch directory: /home/thideras/Folding/CPU
Executable: ./fah6
Arguments: -smp -verbosity 9 

[23:20:49] - Ask before connecting: No
[23:20:49] - User name: thideras (Team 32)
[23:20:49] - User ID: 73C298CA6E2289A0
[23:20:49] - Machine ID: 1
[23:20:49] 
[23:20:49] Loaded queue successfully.
[23:20:49] - Preparing to get new work unit...
[23:20:49] - Autosending finished units...
[23:20:49] Trying to send all finished work units
[23:20:49] + No unsent completed units remaining.
[23:20:49] - Autosend completed
[23:20:49] + Attempting to get work packet
[23:20:49] - Will indicate memory of 1 MB
[23:20:49] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 7, Stepping: 6
[23:20:49] - Connecting to assignment server
[23:20:49] Connecting to http://assign.stanford.edu:8080/
[23:20:50] Posted data.
[23:20:50] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[23:20:50] + Couldn't get work instructions.
[23:20:50] - Attempt #1  to get work failed, and no other work to do.
             Waiting before retry.
[23:21:05] + Attempting to get work packet
[23:21:05] - Will indicate memory of 1 MB
[23:21:05] - Connecting to assignment server
[23:21:05] Connecting to http://assign.stanford.edu:8080/
[23:21:05] Posted data.
[23:21:05] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[23:21:05] + Couldn't get work instructions.
[23:21:05] - Attempt #2  to get work failed, and no other work to do.
             Waiting before retry.
[23:21:28] + Attempting to get work packet
[23:21:28] - Will indicate memory of 1 MB
[23:21:28] - Connecting to assignment server
[23:21:28] Connecting to http://assign.stanford.edu:8080/
[23:21:28] Posted data.
[23:21:28] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[23:21:28] + Couldn't get work instructions.
[23:21:28] - Attempt #3  to get work failed, and no other work to do.
             Waiting before retry.
If I run it without the "-smp" flag, it pulls a single core unit without any issues. I've tried stopping the iptables service to make sure it isn't something on this computer stopping it, didn't help. I've cleared everything out of the folder except for the executable files, same thing.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

For the GPU, I followed our article on the front page and had to make some adjustments for changes since then. Everything reports that it is the right version and that the files are linked, unless I'm blind. When I launch, it says it can't find one of the file and promptly crashes.

Code:
[[email protected] GPU1]$ nice -n 19 wine [email protected] -forcegpu nvidia_g80

Note: Please read the license agreement ([email protected] -license). Further 
use of this software requires that you have read and accepted this agreement.



--- Opening Log file [February 15 23:26:31 UTC] 


# Windows GPU Console Edition #################################################
###############################################################################

                       [email protected] Client Version 6.23

                          http://folding.stanford.edu

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

Launch directory: Z:\home\thideras\Folding\GPU1
Executable: Z:\home\thideras\Folding\GPU1\[email protected]
Arguments: -forcegpu nvidia_g80 

[23:26:31] - Ask before connecting: No
[23:26:31] - User name: thideras (Team 32)
[23:26:31] - User ID: 1134E69E155A280C
[23:26:31] - Machine ID: 2
[23:26:31] 
[23:26:32] Loaded queue successfully.
[23:26:32] 
[23:26:32] + Processing work unit
[23:26:32] Core required: FahCore_11.exe
[23:26:32] Core found.
[23:26:32] Working on queue slot 01 [February 15 23:26:32 UTC]
[23:26:32] + Working ...
err:module:import_dll Library nvcuda.dll (which is needed by L"Z:\\home\\thideras\\Folding\\GPU1\\FahCore_11.exe") not found
err:module:LdrInitializeThunk Main exe initialization for L"Z:\\home\\thideras\\Folding\\GPU1\\FahCore_11.exe" failed, status c0000135
[23:26:36] CoreStatus = C0000135 (-1073741515)
[23:26:36] Client-core communications error: ERROR 0xc0000135
[23:26:36] This is a sign of more serious problems, shutting down.
Code:
[[email protected] GPU1]$ ldd /usr/local/lib/wine/nvcuda.dll.so
    linux-gate.so.1 =>  (0x00f29000)
    libcudart.so.2 => /usr/local/cuda/lib/libcudart.so.2 (0x00afb000)
    libwine.so.1 => /usr/lib/libwine.so.1 (0x00c59000)
    libm.so.6 => /lib/libm.so.6 (0x00242000)
    libc.so.6 => /lib/libc.so.6 (0x0026c000)
    libdl.so.2 => /lib/libdl.so.2 (0x00110000)
    libpthread.so.0 => /lib/libpthread.so.0 (0x008b0000)
    librt.so.1 => /lib/librt.so.1 (0x00723000)
    libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x00115000)
    libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x00f35000)
    /lib/ld-linux.so.2 (0x433ec000)
Ideas? :shrug:
 

Zerix01

Member
Joined
Mar 12, 2007
A few things about the GPU client. Are you running GPU2 or 3? Which Nvidia driver are you using?

That guide seems a little old and it is for GPU2. GPU3 needs the Cuda toolkit 3 and is sensitive to which driver you run with it. The latest and greatest does not work with the current GPU3 wrapper for Cuda 3.

Check this out and pay attention to the versions being used for everything.
http://linuxfah.info/index.php?title=Main_Page

I personally had to stop folding on my two GPU's because they both needed a CPU and running my x6 with SMP was netting a lot more points than SMP on four cores plus the two GPU which also seemed to slow down the SMP client even more than just dropping the two cores alone did.

If you really want to fold on a GPU and SMP and get decent points, then you should install Ubuntu 8.04. Something about the kernel was different and a lot of the current problems do not exist.
 

ChasR

Senior Member
Joined
Apr 12, 2004
Location
Atlanta
You've been doing origami far too long!

Base values of SMP WUs were reduced well over a year ago and the Quick Return Bonus instituted. You have to have a passkey to get the bonus. If you've never used one, you'll have to fold 10 a3 core WUs to qualify for the bonus. You also have to maintain a 80% sucessful and timely completion ratio.

6.02 won't do smp, hence the "no appropriate work server" message. You need 6.29 or higher. You may have to install nscd to make folding work. I posted a how to somewhere. Search for nscd with me as author. You would have to have it to make ubuntu 10.04 or higher work.

GPU folding on Linux kernels higher than 2.6.28 (in ubuntu that's 9.04) can be made to work, but the net gain when running with smp is negligible. In other words the gpu hammers smp production. Forget about it with your setup.

Production from a Q9550 @ 3.6 in an Ubuntu 10.10 VM:

Name: Val-VM 10.10
Path: \\VAL-UBUNTU\fah\
Number of Frames Observed: 114

Min. Time / Frame : 00:04:17 - 10,371 PPD
Avg. Time / Frame : 00:04:22 - 10,076 PPD
Cur. Time / Frame : 00:04:31 - 9,531 PPD
R3F. Time / Frame : 00:04:31 - 9,531 PPD
All Time / Frame : 00:04:31 - 9,531 PPD
Eff. Time / Frame : 00:04:50 - 8,906 PPD

The variations are due to the two gpu clients running in windows.
 
OP
Automata

Automata

Destroyer of Empires and Use
Joined
May 15, 2006
I think I'll skip GPU folding since I probably won't be using a high end GPU in my next build. I don't game anymore.

I installed NSCD, configured it like in your post here and corrected the version I had on my system. Now I'm back to square one:

Code:
# Linux SMP Console Edition ###################################################
###############################################################################

                       [email protected] Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: /home/thideras/Folding/CPU
Executable: ./fah6
Arguments: -smp -verbosity 9 

[03:37:26] - Ask before connecting: No
[03:37:26] - User name: thideras (Team 32)
[03:37:26] - User ID not found locally
[03:37:26] + Requesting User ID from server
[03:37:26] - Getting ID from AS: 
[03:37:26] Connecting to http://assign.stanford.edu:8080/
[03:37:26] - Could not CosmHTTPOpen
[03:37:26] + Could not connect to Primary Assignment Server for ID
[03:37:26] Connecting to http://assign2.stanford.edu:80/
[03:37:26] - Could not CosmHTTPOpen
[03:37:26] + Could not connect to Secondary Assignment Server for ID
[03:37:26] 
+ Could not get ID from server. Retrying...
 
Last edited:
OP
Automata

Automata

Destroyer of Empires and Use
Joined
May 15, 2006
Yup, that is one of the first threads I found regarding this issue. Did everything that was suggested with nothing working.

The computer definitely has a working internet connection. I'm doing Rosetta in the mean time and making this post from said computer. Using the alternate executable file gets past that error, but doesn't pull an assignment.

Code:
[[email protected] CPU]$ ./fah6

Note: Please read the license agreement (fah6 -license). Further 
use of this software requires that you have read and accepted this agreement.

4 cores detected


--- Opening Log file [February 16 04:07:11 UTC] 


# Linux SMP Console Edition ###################################################
###############################################################################

                       [email protected] Client Version 6.24beta

                          http://folding.stanford.edu

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

Launch directory: /home/thideras/Folding/CPU
Executable: ./fah6
Arguments: -smp -verbosity 9 

[04:07:11] - Ask before connecting: No
[04:07:11] - User name: thideras (Team 32)
[04:07:11] - User ID: 4BDF897F3C25EC8F
[04:07:11] - Machine ID: 1
[04:07:11] 
[04:07:11] Loaded queue successfully.
[04:07:11] - Preparing to get new work unit...
[04:07:11] + Attempting to get work packet
[04:07:11] - Will indicate memory of 8001 MB
[04:07:11] - Connecting to assignment server
[04:07:11] Connecting to http://assign.stanford.edu:8080/
[04:07:11] - Autosending finished units... [February 16 04:07:11 UTC]
[04:07:11] Trying to send all finished work units
[04:07:11] + No unsent completed units remaining.
[04:07:11] - Autosend completed
[04:07:11] Posted data.
[04:07:11] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[04:07:11] + Couldn't get work instructions.
[04:07:11] - Attempt #1  to get work failed, and no other work to do.
Waiting before retry.
[04:07:24] + Attempting to get work packet
[04:07:24] - Will indicate memory of 8001 MB
[04:07:24] - Connecting to assignment server
[04:07:24] Connecting to http://assign.stanford.edu:8080/
[04:07:24] Posted data.
[04:07:24] Initial: 0000; + No appropriate work server was available; will try again in a bit.
[04:07:24] + Couldn't get work instructions.
[04:07:24] - Attempt #2  to get work failed, and no other work to do.
Waiting before retry.
^C[04:07:36] ***** Got an Activate signal (2)
[04:07:36] Killing all core threads

[email protected] Client Shutdown.

EDIT: I see your second post, reading...
 
OP
Automata

Automata

Destroyer of Empires and Use
Joined
May 15, 2006
The URL is pretty funny, but it did not help. :-/