large SMP WU upload times

Moderators: Site Moderators, FAHC Science Team

Post Reply
PantherX
Site Moderator
Posts: 6986
Joined: Wed Dec 23, 2009 9:33 am
Hardware configuration: V7.6.21 -> Multi-purpose 24/7
Windows 10 64-bit
CPU:2/3/4/6 -> Intel i7-6700K
GPU:1 -> Nvidia GTX 1080 Ti
§
Retired:
2x Nvidia GTX 1070
Nvidia GTX 675M
Nvidia GTX 660 Ti
Nvidia GTX 650 SC
Nvidia GTX 260 896 MB SOC
Nvidia 9600GT 1 GB OC
Nvidia 9500M GS
Nvidia 8800GTS 320 MB

Intel Core i7-860
Intel Core i7-3840QM
Intel i3-3240
Intel Core 2 Duo E8200
Intel Core 2 Duo E6550
Intel Core 2 Duo T8300
Intel Pentium E5500
Intel Pentium E5400
Location: Land Of The Long White Cloud
Contact:
Contact PantherX

Re: large SMP WU upload times

Post by PantherX »

Here is mine:

Code: Select all


--- Opening Log file [June 27 18:56:23 UTC] 


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

                       Folding@Home Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: D:\F@H SMP
Executable: D:\F@H SMP\Folding@home-Win32-x86.exe
Arguments: -smp 7 -advmethods -verbosity 9 

[18:56:23] - Ask before connecting: No
[18:56:23] - User name: PantherX (Team 69411)
[18:56:23] - User ID: 
[18:56:23] - Machine ID: 1
[18:56:23] 
[18:56:23] Loaded queue successfully.
[18:56:23] - Preparing to get new work unit...
[18:56:23] - Autosending finished units... [June 27 18:56:23 UTC]
[18:56:23] Cleaning up work directory
[18:56:23] Trying to send all finished work units
[18:56:23] Project: 6701 (Run 37, Clone 32, Gen 4)


[18:56:23] + Attempting to send results [June 27 18:56:23 UTC]
[18:56:23] - Reading file work/wuresults_04.dat from core
[18:56:23]   (Read 43233208 bytes from disk)
[18:56:23] Connecting to http://171.64.65.56:8080/
[18:56:23] + Attempting to get work packet
[18:56:23] Passkey found
[18:56:23] - Will indicate memory of 4091 MB
[18:56:23] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 14, Stepping: 5
[18:56:23] - Connecting to assignment server
[18:56:23] Connecting to http://assign.stanford.edu:8080/
[18:56:25] Posted data.
[18:56:25] Initial: 40AB; - Successful: assigned to (171.64.65.56).
[18:56:25] + News From Folding@Home: Welcome to Folding@Home
[18:56:25] Loaded queue successfully.
[18:56:25] Connecting to http://171.64.65.56:8080/
[18:56:33] Posted data.
[18:56:33] Initial: 0000; - Receiving payload (expected size: 763488)
[18:57:15] - Downloaded at ~17 kB/s
[18:57:15] - Averaged speed for that direction ~29 kB/s
[18:57:15] + Received work.
[18:57:15] + Closed connections
[18:57:15] 
[18:57:15] + Processing work unit
[18:57:15] Core required: FahCore_a3.exe
[18:57:15] Core found.
[18:57:15] Working on queue slot 05 [June 27 18:57:15 UTC]
[18:57:15] + Working ...
[18:57:15] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 05 -np 7 -nocpulock -checkpoint 15 -verbose -lifeline 6624 -version 629'

[18:57:15] 
[18:57:15] *------------------------------*
[18:57:15] Folding@Home Gromacs SMP Core
[18:57:15] Version 2.22 (Mar 12, 2010)
[18:57:15] 
[18:57:15] Preparing to commence simulation
[18:57:15] - Looking at optimizations...
[18:57:15] - Created dyn
[18:57:15] - Files status OK
[18:57:15] - Expanded 762976 -> 1404481 (decompressed 184.0 percent)
[18:57:15] Called DecompressByteArray: compressed_data_size=762976 data_size=1404481, decompressed_data_size=1404481 diff=0
[18:57:15] - Digital signature verified
[18:57:15] 
[18:57:15] Project: 6701 (Run 83, Clone 30, Gen 7)
[18:57:15] 
[18:57:15] Assembly optimizations on if available.
[18:57:15] Entering M.D.
[18:57:21] Completed 0 out of 2000000 steps  (0%)
[19:08:00] Completed 20000 out of 2000000 steps  (1%)
[19:19:33] Completed 40000 out of 2000000 steps  (2%)
[19:29:40] Completed 60000 out of 2000000 steps  (3%)
[19:39:48] Completed 80000 out of 2000000 steps  (4%)
[19:49:51] Completed 100000 out of 2000000 steps  (5%)
[19:53:15] Posted data.
[19:53:15] Initial: 0000; - Uploaded at ~12 kB/s
[19:53:16] - Averaged speed for that direction ~8 kB/s
[19:53:16] + Results successfully sent
[19:53:16] Thank you for your contribution to Folding@Home.
[19:53:16] + Number of Units Completed: 181

[19:53:18] + Sent 1 of 1 completed units to the server
[19:53:18] - Autosend completed
[19:59:57] Completed 120000 out of 2000000 steps  (6%)
[20:10:02] Completed 140000 out of 2000000 steps  (7%)
and here is another:

Code: Select all

[11:21:05] Completed 1920000 out of 2000000 steps  (96%)
[11:30:35] Completed 1940000 out of 2000000 steps  (97%)
[11:40:07] Completed 1960000 out of 2000000 steps  (98%)
[11:49:39] Completed 1980000 out of 2000000 steps  (99%)
[11:59:11] Completed 2000000 out of 2000000 steps  (100%)
[11:59:11] DynamicWrapper: Finished Work Unit: sleep=10000
[11:59:21] 
[11:59:21] Finished Work Unit:
[11:59:21] - Reading up to 687408 from "work/wudata_07.trr": Read 687408
[11:59:21] trr file hash check passed.
[11:59:21] - Reading up to 42273764 from "work/wudata_07.xtc": Read 42273764
[11:59:22] xtc file hash check passed.
[11:59:22] edr file hash check passed.
[11:59:22] logfile size: 279083
[11:59:22] Leaving Run
[11:59:24] - Writing 43242591 bytes of core data to disk...
[11:59:25]   ... Done.
[11:59:27] - Shutting down core
[11:59:27] 
[11:59:27] Folding@home Core Shutdown: FINISHED_UNIT
[11:59:30] CoreStatus = 64 (100)
[11:59:30] Unit 7 finished with 89 percent of time to deadline remaining.
[11:59:30] Updated performance fraction: 0.911045
[11:59:30] Sending work to server
[11:59:30] Project: 6701 (Run 36, Clone 11, Gen 14)


[11:59:30] + Attempting to send results [June 29 11:59:30 UTC]
[11:59:30] - Reading file work/wuresults_07.dat from core
[11:59:30]   (Read 43242591 bytes from disk)
[11:59:30] Connecting to http://171.64.65.56:8080/
[12:53:05] - Couldn't send HTTP request to server
[12:53:05] + Could not connect to Work Server (results)
[12:53:05]     (171.64.65.56:8080)
[12:53:05] + Retrying using alternative port
[12:53:05] Connecting to http://171.64.65.56:80/
[13:18:33] - Couldn't send HTTP request to server
[13:18:33] + Could not connect to Work Server (results)
[13:18:33]     (171.64.65.56:80)
[13:18:33] - Error: Could not transmit unit 07 (completed June 29) to work server.
[13:18:33] - 1 failed uploads of this unit.
[13:18:33]   Keeping unit 07 in queue.
[13:18:33] Trying to send all finished work units
[13:18:33] Project: 6701 (Run 36, Clone 11, Gen 14)


[13:18:33] + Attempting to send results [June 29 13:18:33 UTC]
[13:18:33] - Reading file work/wuresults_07.dat from core
[13:18:33]   (Read 43242591 bytes from disk)
[13:18:33] Connecting to http://171.64.65.56:8080/
[13:32:00] - Couldn't send HTTP request to server
[13:32:00] + Could not connect to Work Server (results)
[13:32:00]     (171.64.65.56:8080)
[13:32:00] + Retrying using alternative port
[13:32:00] Connecting to http://171.64.65.56:80/
[13:53:18] - Autosending finished units... [June 29 13:53:18 UTC]
[13:53:18] Trying to send all finished work units
[13:53:18] - Already sending work
[13:53:18] + Sent 0 of 1 completed units to the server
[13:53:18] - Autosend completed
[13:59:31] Killing all core threads
[13:59:31] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown at user request.
[13:59:31] ***** Got a SIGTERM signal (2)
[13:59:31] Killing all core threads
[13:59:31] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown.

--- Opening Log file [June 29 13:59:34 UTC] 


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

                       Folding@Home Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: D:\F@H SMP
Executable: D:\F@H SMP\Folding@home-Win32-x86.exe
Arguments: -smp 7 -advmethods -verbosity 9 

[13:59:34] - Ask before connecting: No
[13:59:34] - User name: PantherX (Team 69411)
[13:59:34] - User ID: 
[13:59:34] - Machine ID: 1
[13:59:34] 
[13:59:34] Loaded queue successfully.
[13:59:34] - Preparing to get new work unit...
[13:59:34] - Autosending finished units... [June 29 13:59:34 UTC]
[13:59:34] Cleaning up work directory
[13:59:34] Trying to send all finished work units
[13:59:34] Project: 6701 (Run 36, Clone 11, Gen 14)


[13:59:34] + Attempting to send results [June 29 13:59:34 UTC]
[13:59:34] - Reading file work/wuresults_07.dat from core
[13:59:34]   (Read 43242591 bytes from disk)
[13:59:34] Connecting to http://171.64.65.56:8080/
[13:59:35] + Attempting to get work packet
[13:59:35] Passkey found
[13:59:35] - Will indicate memory of 4091 MB
[13:59:35] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 14, Stepping: 5
[13:59:35] - Connecting to assignment server
[13:59:35] Connecting to http://assign.stanford.edu:8080/
[13:59:40] Posted data.
[13:59:41] Initial: ED82; - Successful: assigned to (130.237.232.140).
[13:59:41] + News From Folding@Home: Welcome to Folding@Home
[13:59:41] Loaded queue successfully.
[13:59:41] Connecting to http://130.237.232.140:8080/
[13:59:49] Posted data.
[13:59:49] Initial: 0000; - Receiving payload (expected size: 1799135)
[14:00:59] - Downloaded at ~25 kB/s
[14:00:59] - Averaged speed for that direction ~31 kB/s
[14:00:59] + Received work.
[14:00:59] + Closed connections
[14:00:59] 
[14:00:59] + Processing work unit
[14:00:59] Core required: FahCore_a3.exe
[14:00:59] Core found.
[14:00:59] Working on queue slot 08 [June 29 14:00:59 UTC]
[14:00:59] + Working ...
[14:00:59] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 08 -np 7 -nocpulock -checkpoint 15 -verbose -lifeline 9204 -version 629'

[14:00:59] 
[14:00:59] *------------------------------*
[14:00:59] Folding@Home Gromacs SMP Core
[14:00:59] Version 2.22 (Mar 12, 2010)
[14:00:59] 
[14:00:59] Preparing to commence simulation
[14:00:59] - Looking at optimizations...
[14:00:59] - Created dyn
[14:00:59] - Files status OK
[14:00:59] - Expanded 1798623 -> 2396877 (decompressed 133.2 percent)
[14:00:59] Called DecompressByteArray: compressed_data_size=1798623 data_size=2396877, decompressed_data_size=2396877 diff=0
[14:00:59] - Digital signature verified
[14:00:59] 
[14:00:59] Project: 6014 (Run 1, Clone 209, Gen 151)
[14:00:59] 
[14:00:59] Assembly optimizations on if available.
[14:00:59] Entering M.D.
[14:01:05] Completed 0 out of 500000 steps  (0%)
[14:04:56] Completed 5000 out of 500000 steps  (1%)
[14:08:46] Completed 10000 out of 500000 steps  (2%)
[14:12:37] Completed 15000 out of 500000 steps  (3%)
[14:16:33] Completed 20000 out of 500000 steps  (4%)
[14:20:58] Completed 25000 out of 500000 steps  (5%)
[14:25:38] Completed 30000 out of 500000 steps  (6%)
[14:30:19] Completed 35000 out of 500000 steps  (7%)
[14:34:56] Completed 40000 out of 500000 steps  (8%)
[14:39:35] Completed 45000 out of 500000 steps  (9%)
[14:44:14] Completed 50000 out of 500000 steps  (10%)
[14:48:55] Completed 55000 out of 500000 steps  (11%)
[14:53:34] Completed 60000 out of 500000 steps  (12%)
[14:57:37] Completed 65000 out of 500000 steps  (13%)
[14:58:13] Posted data.
[14:58:13] Initial: 0000; - Uploaded at ~11 kB/s
[14:58:14] - Averaged speed for that direction ~10 kB/s
[14:58:14] + Results successfully sent
[14:58:14] Thank you for your contribution to Folding@Home.
[14:58:14] + Number of Units Completed: 184

[14:58:15] + Sent 1 of 1 completed units to the server
[14:58:15] - Autosend completed
[15:01:28] Completed 70000 out of 500000 steps  (14%)
[15:05:14] Completed 75000 out of 500000 steps  (15%)
[15:09:00] Completed 80000 out of 500000 steps  (16%)
ETA:
Now ↞ Very Soon ↔ Soon ↔ Soon-ish ↔ Not Soon ↠ End Of Time

Welcome To The F@H Support Forum Ӂ Troubleshooting Bad WUs Ӂ Troubleshooting Server Connectivity Issues
Top
bollix47
Posts: 2959
Joined: Sun Dec 02, 2007 5:04 am
Location: Canada

Re: large SMP WU upload times

Post by bollix47 »

I can reproduce this in Windows or Linux.

The autosend runs every time you start the client and every 6 hours thereafter on all operating systems.

Here's a log where the upload seemed to fail so I quit the client and restarted:

Code: Select all

[08:32:02] + Attempting to send results [July 2 08:32:02 UTC]
[08:32:02] - Reading file work/wuresults_03.dat from core
[08:32:02]   (Read 100020561 bytes from disk)
[08:32:02] Connecting to http://171.67.108.22:8080/
[09:17:34] Killing all core threads
[09:17:34] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown at user request.
[09:17:34] ***** Got a SIGTERM signal (2)
[09:17:34] Killing all core threads
[09:17:34] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown.


--- Opening Log file [July 2 09:17:48 UTC] 


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

                       Folding@Home Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: C:\fah\smp
Executable: Folding@home-Win32-x86.exe
Arguments: -smp 15 -verbosity 9 

[09:17:48] - Ask before connecting: No
[09:17:48] - User name: bollix47 (Team 39340)

[09:17:48] - Machine ID: 3
[09:17:48] 
[09:17:48] Loaded queue successfully.
[09:17:48] - Preparing to get new work unit...
[09:17:48] - Autosending finished units... [July 2 09:17:48 UTC]
[09:17:48] Cleaning up work directory
[09:17:48] Trying to send all finished work units
[09:17:48] Project: 2684 (Run 5, Clone 12, Gen 4)


[09:17:48] + Attempting to send results [July 2 09:17:48 UTC]
[09:17:48] - Reading file work/wuresults_03.dat from core
[09:17:48]   (Read 100020561 bytes from disk)
[09:17:48] Connecting to http://171.67.108.22:8080/
[09:17:48] + Attempting to get work packet
[09:17:48] Passkey found
[09:17:48] - Will indicate memory of 12279 MB
[09:17:48] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 5
[09:17:48] - Connecting to assignment server
[09:17:48] Connecting to http://assign.stanford.edu:8080/
[09:17:51] Posted data.
[09:17:51] Initial: ED82; - Successful: assigned to (130.237.232.140).
[09:17:51] + News From Folding@Home: Welcome to Folding@Home
[09:17:51] Loaded queue successfully.
[09:17:51] Connecting to http://130.237.232.140:8080/
[09:17:52] Posted data.
[09:17:52] Initial: 0000; - Receiving payload (expected size: 1798323)
[09:17:55] - Downloaded at ~585 kB/s
[09:17:55] - Averaged speed for that direction ~677 kB/s
[09:17:55] + Received work.
[09:17:55] + Closed connections
[09:17:55] 
[09:17:55] + Processing work unit
[09:17:55] Core required: FahCore_a3.exe
[09:17:55] Core found.
[09:17:55] Working on queue slot 04 [July 2 09:17:55 UTC]
[09:17:55] + Working ...
[09:17:55] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 04 -np 15 -checkpoint 15 -verbose -lifeline 4392 -version 629'

[09:17:55] 
[09:17:55] *------------------------------*
[09:17:55] Folding@Home Gromacs SMP Core
[09:17:55] Version 2.22 (Mar 12, 2010)
[09:17:55] 
[09:17:55] Preparing to commence simulation
[09:17:55] - Looking at optimizations...
[09:17:55] - Created dyn
[09:17:55] - Files status OK
[09:17:56] - Expanded 1797811 -> 2392545 (decompressed 133.0 percent)
[09:17:56] Called DecompressByteArray: compressed_data_size=1797811 data_size=2392545, decompressed_data_size=2392545 diff=0
[09:17:56] - Digital signature verified
[09:17:56] 
[09:17:56] Project: 6015 (Run 0, Clone 38, Gen 205)
[09:17:56] 
[09:17:56] Assembly optimizations on if available.
[09:17:56] Entering M.D.
[09:18:03] Completed 0 out of 500000 steps  (0%)
[09:24:16] - Couldn't send HTTP request to server
[09:24:16] + Could not connect to Work Server (results)
[09:24:16]     (171.67.108.22:8080)
[09:24:16] + Retrying using alternative port
[09:24:16] Connecting to http://171.67.108.22:80/
[09:26:30] Completed 5000 out of 500000 steps  (1%)
[09:29:01] Completed 10000 out of 500000 steps  (2%)
[09:31:32] Completed 15000 out of 500000 steps  (3%)
[09:34:00] Completed 20000 out of 500000 steps  (4%)
[09:36:28] Completed 25000 out of 500000 steps  (5%)
[09:37:45] Posted data.
[09:37:45] Initial: 0000; + Results successfully sent
[09:37:46] Thank you for your contribution to Folding@Home.
[09:37:46] + Number of Units Completed: 245

[09:37:48] + Sent 1 of 1 completed units to the server
[09:37:48] - Autosend completed
Image
Top
tear
Posts: 254
Joined: Sun Dec 02, 2007 4:08 am
Hardware configuration: None
Location: Rocky Mountains

Re: large SMP WU upload times

Post by tear »

Ah, nevermind. I've mistaken it for another behaviour. Pardon me.
One man's ceiling is another man's floor.
Image
Top
gwildperson
Posts: 450
Joined: Tue Dec 04, 2007 8:36 pm

Re: large SMP WU upload times

Post by gwildperson »

tear wrote:Interesting, care to post the log?

Linux client doesn't return a unit from autosend context until a number of
failures has occurred. Perhaps Windows client does things differently (so
much for a unified codebase)?
I assume you're able to sort out which messages come from the download and WU processing and which come from the upload.

Code: Select all

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

                       Folding@Home Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: C:\FAH
Executable: C:\FAH\Folding@home-Win32-x86.exe
Arguments: -verbosity 9 -smp -advmethods -forceasm 

[13:59:56] - Ask before connecting: No
[13:59:56] - User name: ****************
[13:59:56] - User ID: ********
[13:59:56] - Machine ID: 1
[13:59:56] 
[13:59:56] Loaded queue successfully.
[13:59:56] - Preparing to get new work unit...
[13:59:56] - Autosending finished units... [July 6 13:59:56 UTC]
[13:59:56] Cleaning up work directory
[13:59:56] Trying to send all finished work units
[13:59:56] + Attempting to get work packet
[13:59:56] Project: 6015 (Run 1, Clone 164, Gen 157)
[13:59:56] Passkey found


[13:59:56] - Will indicate memory of 8157 MB
[13:59:56] + Attempting to send results [July 6 13:59:56 UTC]
[13:59:56] - Detect CPU.[13:59:56] - Reading file work/wuresults_08.dat from core
 Vendor: GenuineIntel, Family: 6, Model: 7, Stepping: 10
[13:59:56] - Connecting to assignment server
[13:59:56] Connecting to http://assign.stanford.edu:8080/
[13:59:56]   (Read 20542540 bytes from disk)
[13:59:56] Connecting to http://130.237.232.140:8080/
[13:59:56] Posted data.
[13:59:56] Initial: 40AB; - Successful: assigned to (171.64.65.54).
[13:59:56] + News From Folding@Home: Welcome to Folding@Home
[13:59:56] Loaded queue successfully.
[13:59:56] Sent data
[13:59:56] Connecting to http://171.64.65.54:8080/
[13:59:57] Posted data.
[13:59:57] Initial: 0000; - Receiving payload (expected size: 1778386)
[14:01:38] - Downloaded at ~17 kB/s
[14:01:38] - Averaged speed for that direction ~85 kB/s
[14:01:38] + Received work.
[14:01:38] + Closed connections
[14:01:38] 
[14:01:38] + Processing work unit
[14:01:38] Core required: FahCore_a3.exe
[14:01:38] Core found.
[14:01:38] Working on queue slot 09 [July 6 14:01:38 UTC]
[14:01:38] + Working ...
[14:01:38] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 09 -np 4 -priority 96 -nocpulock -checkpoint 15 -forceasm -verbose -lifeline 2180 -version 630'

[14:01:38] 
[14:01:38] *------------------------------*
[14:01:38] Folding@Home Gromacs SMP Core
[14:01:38] Version 2.19 (Mar 12, 2010)
[14:01:38] 
[14:01:38] Preparing to commence simulation
[14:01:38] - Assembly optimizations manually forced on.
[14:01:38] - Not checking prior termination.
[14:01:39] - Expanded 1777874 -> 2058129 (decompressed 115.7 percent)
[14:01:39] Called DecompressByteArray: compressed_data_size=1777874 data_size=2058129, decompressed_data_size=2058129 diff=0
[14:01:39] - Digital signature verified
[14:01:39] 
[14:01:39] Project: 6025 (Run 1, Clone 158, Gen 165)
[14:01:39] 
[14:01:39] Assembly optimizations on if available.
[14:01:39] Entering M.D.
[14:01:46] Completed 0 out of 500000 steps  (0%)
[14:08:40] Posted data.
[14:08:41] Initial: 0000; - Uploaded at ~38 kB/s
[14:08:41] - Averaged speed for that direction ~31 kB/s
[14:08:41] + Results successfully sent
[14:08:41] Thank you for your contribution to Folding@Home.
[14:08:41] + Number of Units Completed: 20

[14:08:46] + Sent 1 of 1 completed units to the server
[14:08:46] - Autosend completed
[14:09:25] Completed 5000 out of 500000 steps  (1%)
Top
theteofscuba
Posts: 96
Joined: Wed Dec 05, 2007 7:15 am
Hardware configuration: PS3, Phenom II X4, QX9775, HD 8570
Contact:
Contact theteofscuba

Re: large SMP WU upload times

Post by theteofscuba »

okay.

there is still one thing: the v6 client does not start folding the new WU while the finished WU is uploading. that's kind of the reason for wanting to download a new WU before uploading a finished WU.
Top
codysluder
Posts: 1024
Joined: Sun Dec 02, 2007 12:43 pm

Re: large SMP WU upload times

Post by codysluder »

theteofscuba wrote:okay.

there is still one thing: the v6 client does not start folding the new WU while the finished WU is uploading. that's kind of the reason for wanting to download a new WU before uploading a finished WU.
That depends on the relative sizes of the upload and download and the relative speeds of your connection uploading and downloading. Take a look at the log posted just above your post.

[14:01:46] Completed 0 out of 500000 steps (0%) --- Initialization complete; processing actually started
[14:08:40] Posted data. --- upload actually finished.
Top
theteofscuba
Posts: 96
Joined: Wed Dec 05, 2007 7:15 am
Hardware configuration: PS3, Phenom II X4, QX9775, HD 8570
Contact:
Contact theteofscuba

Re: large SMP WU upload times

Post by theteofscuba »

I assure you it is not working as I intend it to. the time stamps clearly show that the work unit does not begin until ~ 18 mins after FINISHED_UNIT. 18 minutes it took to upload this SMP WU and the cpu sat mostly idle.

Code: Select all

03:35:06] Completed 1860000 out of 2000000 steps  (93%)
[03:42:17] Completed 1880000 out of 2000000 steps  (94%)
[03:49:26] Completed 1900000 out of 2000000 steps  (95%)
[03:56:38] Completed 1920000 out of 2000000 steps  (96%)
[04:03:48] Completed 1940000 out of 2000000 steps  (97%)
[04:11:09] Completed 1960000 out of 2000000 steps  (98%)
[04:18:20] Completed 1980000 out of 2000000 steps  (99%)
[04:25:28] Completed 2000000 out of 2000000 steps  (100%)
[04:25:29] DynamicWrapper: Finished Work Unit: sleep=10000
[04:25:39] 
[04:25:39] Finished Work Unit:
[04:25:39] - Reading up to 687408 from "work/wudata_05.trr": Read 687408
[04:25:39] trr file hash check passed.
[04:25:39] - Reading up to 42655836 from "work/wudata_05.xtc": Read 42655836
[04:25:39] xtc file hash check passed.
[04:25:39] edr file hash check passed.
[04:25:39] logfile size: 278790
[04:25:39] Leaving Run
[04:25:43] - Writing 43624370 bytes of core data to disk...
[04:25:45]   ... Done.
[04:25:52] - Shutting down core
[04:25:52] 
[04:25:52] Folding@home Core Shutdown: FINISHED_UNIT
[04:25:55] CoreStatus = 64 (100)
[04:25:55] Unit 5 finished with 84 percent of time to deadline remaining.
[04:25:55] Updated performance fraction: 0.838548
[04:25:55] Sending work to server
[04:25:55] Project: 6701 (Run 87, Clone 26, Gen 16)


[04:25:55] + Attempting to send results [July 9 04:25:55 UTC]
[04:25:55] - Reading file work/wuresults_05.dat from core
[04:25:55]   (Read 43624370 bytes from disk)
[04:25:55] Connecting to http://171.64.65.56:8080/
[04:44:17] Posted data.
[04:44:17] Initial: 0000; - Uploaded at ~38 kB/s
[04:44:18] - Averaged speed for that direction ~39 kB/s
[04:44:18] + Results successfully sent
[04:44:18] Thank you for your contribution to Folding@Home.
[04:44:18] + Number of Units Completed: 4

[04:44:24] Trying to send all finished work units
[04:44:24] + No unsent completed units remaining.
[04:44:24] - Preparing to get new work unit...
[04:44:24] Cleaning up work directory
[04:44:24] + Attempting to get work packet
[04:44:24] Passkey found
[04:44:24] - Will indicate memory of 2044 MB
[04:44:24] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 7, Stepping: 6
[04:44:24] - Connecting to assignment server
[04:44:24] Connecting to http://assign.stanford.edu:8080/
[04:44:24] Posted data.
[04:44:24] Initial: 40AB; - Successful: assigned to (171.64.65.56).
[04:44:24] + News From Folding@Home: Welcome to Folding@Home
[04:44:24] Loaded queue successfully.
[04:44:24] Sent data
[04:44:24] Connecting to http://171.64.65.56:8080/
[04:44:25] Posted data.
[04:44:25] Initial: 0000; - Receiving payload (expected size: 764434)
[04:44:29] - Downloaded at ~186 kB/s
[04:44:29] - Averaged speed for that direction ~256 kB/s
[04:44:29] + Received work.
[04:44:29] Trying to send all finished work units
[04:44:29] + No unsent completed units remaining.
[04:44:29] + Closed connections
[04:44:29] 
[04:44:29] + Processing work unit
[04:44:29] Core required: FahCore_a3.exe
[04:44:29] Core found.
[04:44:29] Working on queue slot 06 [July 9 04:44:29 UTC]
[04:44:29] + Working ...
[04:44:29] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 06 -np 8 -priority 96 -checkpoint 15 -service -verbose -lifeline 2396 -version 630'

[04:44:29] 
[04:44:29] *------------------------------*
[04:44:29] Folding@Home Gromacs SMP Core
[04:44:29] Version 2.22 (Mar 12, 2010)
[04:44:29] 
[04:44:29] Preparing to commence simulation
[04:44:29] - Looking at optimizations...
[04:44:29] - Created dyn
[04:44:29] - Files status OK
[04:44:29] - Expanded 763922 -> 1404481 (decompressed 183.8 percent)
[04:44:29] Called DecompressByteArray: compressed_data_size=763922 data_size=1404481, decompressed_data_size=1404481 diff=0
[04:44:29] - Digital signature verified
[04:44:29] 
[04:44:29] Project: 6702 (Run 4, Clone 51, Gen 2)
[04:44:29] 
[04:44:29] Assembly optimizations on if available.
[04:44:29] Entering M.D.
[04:44:35] Completed 0 out of 2000000 steps  (0%)
[04:51:49] Completed 20000 out of 2000000 steps  (1%)
[04:59:13] Completed 40000 out of 2000000 steps  (2%)
[05:06:27] Completed 60000 out of 2000000 steps  (3%)
[05:13:41] Completed 80000 out of 2000000 steps  (4%)
[05:20:55] Completed 100000 out of 2000000 steps  (5%)
[05:28:09] Completed 120000 out of 2000000 steps  (6%)
Top
codysluder
Posts: 1024
Joined: Sun Dec 02, 2007 12:43 pm

Re: large SMP WU upload times

Post by codysluder »

You don't seem to understand what has been said.

Certainly the behavior between WUs needs to be improved. That's the original topic. The behavior after a restart is much better. The suggestion is to improve v7 so that the behavior between WUs is the same as at the beginning.

If you happen to be around when your client finishes a WU which will take hours to upload, restart the client and save those hours. Be sure to wait until the log says
[19:07:09] CoreStatus = 64 (100)
[19:07:09] Sending work to server
Top
kiore
Posts: 921
Joined: Fri Jan 16, 2009 5:45 pm
Location: USA

Re: large SMP WU upload times

Post by kiore »

codysluder wrote:You don't seem to understand what has been said.

Certainly the behavior between WUs needs to be improved. That's the original topic. The behavior after a restart is much better. The suggestion is to improve v7 so that the behavior between WUs is the same as at the beginning.

If you happen to be around when your client finishes a WU which will take hours to upload, restart the client and save those hours. Be sure to wait until the log says
[19:07:09] CoreStatus = 64 (100)
[19:07:09] Sending work to server
I agree, I have a very slow connection and if I get a "cannot connect to server" I restart, usually I see that the new unit downloaded but did not start an hour or so before and restarting gets it underway. The successful send message usually then occurs in the first 4-6% of the new unit. I think I'll just restart all that I'm around for as I lose quite a few hours a day otherwise. Hopefully as the client matures this will not be such an issue, either that or my internet connect speed improves.. :roll: But I think I'll be setting up a separate internet connection for my next system, as 2 systems running bigadv and 1 running smps is going to be a serious bottle neck and the 2 modems I have now will just not make it.
Image
i7 7800x RTX 3070 OS= win10. AMD 3700x RTX 2080ti OS= win10 .

Team page: https://www.rationalskepticism.org/viewtopic.php?t=616
Top
tear
Posts: 254
Joined: Sun Dec 02, 2007 4:08 am
Hardware configuration: None
Location: Rocky Mountains

Re: large SMP WU upload times

Post by tear »

Please see Langouste Windows port alpha test announcement -- viewtopic.php?f=14&t=15250

Thanks,
tear
One man's ceiling is another man's floor.
Image
Top
tear
Posts: 254
Joined: Sun Dec 02, 2007 4:08 am
Hardware configuration: None
Location: Rocky Mountains

Re: large SMP WU upload times

Post by tear »

Released, also with upload capping feature. Go there.
One man's ceiling is another man's floor.
Image
Top
Post Reply

Return to “Discussions of General-FAH topics”