Trying my hand at SMP

[bubba]

Well waited it out, came in this morning and looked at the box and it was folding away. Went and looked through the log and looks like after the 3rd atempt it downloaded and started the next unit.

Code: Select all

[12:30:48] + Attempting to send results [July 25 12:30:48 UTC]
[12:30:48] - Reading file work/wuresults_03.dat from core
[12:30:48]   (Read 21934933 bytes from disk)
[12:30:48] Connecting to http://171.64.65.64:8080/
[12:50:48] Posted data.
[13:10:48] Initial: 0000; + Could not connect to Work Server (results)
[13:30:48]     (171.64.65.64:8080)
[13:30:48] + Retrying using alternative port
[13:30:48] Connecting to http://171.64.65.64:80/
[13:39:45] - Autosending finished units... [July 25 13:39:45 UTC]
[13:39:45] Trying to send all finished work units
[13:39:45] - Already sending work
[13:39:45] + Sent 0 of 1 completed units to the server
[13:39:45] - Autosend completed
[13:50:49] Posted data.
[14:10:49] Initial: 0000; + Could not connect to Work Server (results)
[14:30:49]     (171.64.65.64:80)
[14:30:49] - Error: Could not transmit unit 03 (completed July 25) to work server.
[14:30:49] - 3 failed uploads of this unit.


[14:30:49] + Attempting to send results [July 25 14:30:49 UTC]
[14:30:49] - Reading file work/wuresults_03.dat from core
[14:30:49]   (Read 21934933 bytes from disk)
[14:30:49] Connecting to http://171.64.122.86:8080/
[14:50:49] Posted data.
[15:10:49] Initial: 0000; + Could not connect to Work Server (results)
[15:30:49]     (171.64.122.86:8080)
[15:30:49] + Retrying using alternative port
[15:30:49] Connecting to http://171.64.122.86:80/
[15:30:49] - Couldn't send HTTP request to server
[15:30:49] + Could not connect to Work Server (results)
[15:30:49]     (171.64.122.86:80)
[15:30:49]   Could not transmit unit 03 to Collection server; keeping in queue.
[15:30:49] + Sent 0 of 1 completed units to the server
[15:30:49] + Closed connections
[15:30:49] 
[15:30:49] + Processing work unit
[15:30:49] Work type a1 not eligible for variable processors
[15:30:49] Core required: FahCore_a1.exe
[15:30:49] Core found.
[15:30:49] Using generic mpiexec calls
[15:30:49] Working on queue slot 04 [July 25 15:30:49 UTC]
[15:30:49] + Working ...
[15:30:49] - Calling 'mpiexec -np 4 -channel auto -host 127.0.0.1 FahCore_a1.exe -dir work/ -suffix 04 -checkpoint 15 -verbose -lifeline 1528 -version 622'

[15:30:50] 
[15:30:50] *------------------------------*
[15:30:50] Folding@Home Gromacs SMP Core
[15:30:50] Version 1.74 (March 10, 2007)
[15:30:50] 
[15:30:50] Preparing to commence simulation
[15:30:50] - Ensuring status. Please wait.
[15:30:55] - Starting from initial work packet
[15:30:56] 
[15:30:56] Project: 2665 (Run 2, Clone 32, Gen 38)
[15:30:56] 
[15:30:56] Assembly optimizations on if available.
[15:30:56] Entering M.D.
[15:31:19] ial work pa- Starting from initial work packet
[15:31:19] 
[15:31:19] Project: 2665 (Run 2, Clone 32, Gen 38)
[15:31:19] 
[15:31:21] Entering M.D.
[15:31:27] Rejecting checkpoint
[15:31:28] Protein: HGG with glycosylations
[15:31:28] Writing local files
[15:31:36] Extra SSE boost OK.
[15:31:36] Writing local files
[15:31:36] Completed 0 out of 250000 steps  (0 percent)
[15:46:25] Writing local files
[15:46:25] Completed 2500 out of 250000 steps  (1 percent)
[16:01:12] Writing local files
[16:01:12] Completed 5000 out of 250000 steps  (2 percent)

Then later it tried again and was finaly able to send it.

Code: Select all

[19:13:16] Completed 37500 out of 250000 steps  (15 percent)
[19:28:02] Writing local files
[19:28:03] Completed 40000 out of 250000 steps  (16 percent)
[19:39:45] - Autosending finished units... [July 25 19:39:45 UTC]
[19:39:45] Trying to send all finished work units
[19:39:45] Project: 2665 (Run 1, Clone 989, Gen 39)


[19:39:45] + Attempting to send results [July 25 19:39:45 UTC]
[19:39:45] - Reading file work/wuresults_03.dat from core
[19:39:45]   (Read 21934933 bytes from disk)
[19:39:45] Connecting to http://171.64.65.64:8080/
[19:42:50] Writing local files
[19:42:51] Completed 42500 out of 250000 steps  (17 percent)
[19:43:49] Posted data.
[19:43:49] Initial: 0000; - Uploaded at ~87 kB/s
[19:43:50] - Averaged speed for that direction ~97 kB/s
[19:43:50] + Results successfully sent
[19:43:50] Thank you for your contribution to Folding@Home.
[19:43:50] + Number of Units Completed: 3

[19:43:50] + Sent 1 of 1 completed units to the server
[19:43:50] - Autosend completed
[19:57:39] Writing local files
[19:57:39] Completed 45000 out of 250000 steps  (18 percent)
[20:12:26] Writing local files
[20:12:26] Completed 47500 out of 250000 steps  (19 percent)

Next time I will wait it out at least a day to see what happens.

[DMB2000uk]

That's good news to hear.

The folding clients are pretty good when it comes to trying to get themselves working again.

Dan

[XstollieX]

you want -smp -verbosity 9 -forceasm -advmethods

What do the -smp and the -verbosity 9 flags do? When i was running the 5.92 i didn't have any flags set up

[DMB2000uk]

Check out the wiki for the full listings of flags

-verbosity x
Sets the output level, from 1 to 9 (max). The default is 3. This option should be used if the user wishes to see more detailed client output, particularly useful if there is a problem which needs to be reported to the F@H client development team.

Some of the new clients don't automatically run SMP (multicore) WU's and default to the single core WUs, which obviously isn't using the CPU to its full potential:

-smp

Sets the universal client v6 beta for Linux to run SMP work units. The client will default to run CPU client work units without this switch.

Dan