need help setting up bigadv under linux

4

404 - File not found

Gawd
Joined
Apr 2, 2007
Messages
525
Any ideas why I'm getting poor performance?

Log:
16 cores detected


--- Opening Log file [April 1 04:22:18 UTC]


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

Folding@Home Client Version 6.29

http://folding.stanford.edu

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

Launch directory: /home/remi/folding
Executable: ./fah6
Arguments: -smp 16 -bigadv -verbosity 9

[04:22:18] - Ask before connecting: No
[04:22:18] - User name: 404_-_File_not_found (Team 33)
[04:22:18] - User ID: 34771243568034E8
[04:22:18] - Machine ID: 1


[04:26:58] Trying to unzip core FahCore_a2.exe
[04:26:58] Decompressed FahCore_a2.exe (5509624 bytes) successfully
[04:26:58] + Core successfully engaged
[04:27:36]
[04:27:36] + Processing work unit
[04:27:36] Core required: FahCore_a2.exe
[04:27:36] Core found.
[04:27:36] Working on queue slot 01 [April 1 04:27:36 UTC]
[04:27:36] + Working ...
[04:27:36] - Calling './mpiexec -np 16 -host 127.0.0.1 ./FahCore_a2.exe -dir work/ -nice 19 -suffix 01 -checkpoint 15 -verbose -lifeline 7604 -version 629'

[04:27:36]
[04:27:36] *------------------------------*
[04:27:36] Folding@Home Gromacs SMP Core
[04:27:36] Version 2.10 (Sun Aug 30 03:43:28 CEST 2009)
[04:27:36]
[04:27:36] Preparing to commence simulation
[04:27:36] - Ensuring status. Please wait.
[04:27:44] Called DecompressByteArray: compressed_data_size=30240067 data_size=159270593, decompressed_data_size=159270593 diff=0
[04:27:46] - Digital signature verified
[04:27:46]
[04:27:46] Project: 2683 (Run 2, Clone 2, Gen 65)
[04:27:46]
[04:27:46] Assembly optimizations on if available.
[04:27:46] Entering M.D.
[04:27:55] (Run 2, Clone 2, Gen 65)
[04:27:55]
[04:27:56] Entering M.D.


...

Making 2D domain decomposition 4 x 1 x 3
starting mdrun 'SINGLE VESICLE in water'
16500000 steps, 66000.0 ps (continuing from step 16250000, 65000.0 ps).
[05:45:40] pleted 2500 out of 250000 steps (1%)
[06:46:45] Completed 5000 out of 250000 steps (2%)
[07:56:35] Completed 7500 out of 250000 steps (3%)
[09:04:41] Completed 10000 out of 250000 steps (4%)
[10:21:46] Completed 12500 out of 250000 steps (5%)
Click to expand...


Benchmarks
CPU Blowfish
CPU Blowfish
This Machine 1800 MHz 1.983
Intel(R) Celeron(R) M processor 1.50GHz (null) 26.1876862
PowerPC 740/750 (280.00MHz) (null) 172.816713
CPU CryptoHash
CPU Cryptohash
This Machine 1800 MHz 664.695
CPU Fibonacci
CPU Fibonacci
This Machine 1800 MHz 3.719
Intel(R) Celeron(R) M processor 1.50GHz (null) 8.1375674
PowerPC 740/750 (280.00MHz) (null) 58.07682
CPU N-Queens
CPU N-Queens
This Machine 1800 MHz 1.241
FPU FFT
FPU FFT
This Machine 1800 MHz 3.010
FPU Raytracing
FPU Raytracing
This Machine 1800 MHz 26.069
Intel(R) Celeron(R) M processor 1.50GHz (null) 40.8816714
PowerPC 740/750 (280.00MHz) (null) 161.312647
Click to expand...
 
Last edited:
Is it downloading an fah6 file in that directory? I have never tried to use finstall myself. Can you paste the ls output from this directory?

 
what distro are you running?

Looks like you are running a 64-bit distro but don't have 32-bit libraries installed. You need them.
 
I never had luck with finstall.

All I did was DL the fah6.exe from stanford and run it with the proper flags.
 
Pretty much all I did. Ran finstall. Then after it was complete, set the flags for Big Adv and replace the executable (fah6).

Once I created an image, I've just kept replicating it.
 
this is native because I have 16 cores, otherwise I would have went EVGA VM, so much easier.
It's been a pain since the beginning -__-
 
In your client.cfg file:

what do you have under extra parms?

Are you running smp -16? 8GB of RAM on the box correct (based on your sig)?

Yeah, you should be getting decent frame times (in the ~35 to ~40 minute range).

Any way to see your CPU and Memory utilization?
 
yeah memory gets 75% usage, CPU goes from 88% to 99% usage, there's almost always a core who's not used fully, but that's probably normal.
 
Last edited:
Log:
16 cores detected


--- Opening Log file [April 1 04:22:18 UTC]


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

Folding@Home Client Version 6.29

http://folding.stanford.edu

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

Launch directory: /home/remi/folding
Executable: ./fah6
Arguments: -smp 16 -bigadv -verbosity 9

[04:22:18] - Ask before connecting: No
[04:22:18] - User name: 404_-_File_not_found (Team 33)
[04:22:18] - User ID: 34771243568034E8
[04:22:18] - Machine ID: 1


[04:26:58] Trying to unzip core FahCore_a2.exe
[04:26:58] Decompressed FahCore_a2.exe (5509624 bytes) successfully
[04:26:58] + Core successfully engaged
[04:27:36]
[04:27:36] + Processing work unit
[04:27:36] Core required: FahCore_a2.exe
[04:27:36] Core found.
[04:27:36] Working on queue slot 01 [April 1 04:27:36 UTC]
[04:27:36] + Working ...
[04:27:36] - Calling './mpiexec -np 16 -host 127.0.0.1 ./FahCore_a2.exe -dir work/ -nice 19 -suffix 01 -checkpoint 15 -verbose -lifeline 7604 -version 629'

[04:27:36]
[04:27:36] *------------------------------*
[04:27:36] Folding@Home Gromacs SMP Core
[04:27:36] Version 2.10 (Sun Aug 30 03:43:28 CEST 2009)
[04:27:36]
[04:27:36] Preparing to commence simulation
[04:27:36] - Ensuring status. Please wait.
[04:27:44] Called DecompressByteArray: compressed_data_size=30240067 data_size=159270593, decompressed_data_size=159270593 diff=0
[04:27:46] - Digital signature verified
[04:27:46]
[04:27:46] Project: 2683 (Run 2, Clone 2, Gen 65)
[04:27:46]
[04:27:46] Assembly optimizations on if available.
[04:27:46] Entering M.D.
[04:27:55] (Run 2, Clone 2, Gen 65)
[04:27:55]
[04:27:56] Entering M.D.


...

Making 2D domain decomposition 4 x 1 x 3
starting mdrun 'SINGLE VESICLE in water'
16500000 steps, 66000.0 ps (continuing from step 16250000, 65000.0 ps).
[05:45:40] pleted 2500 out of 250000 steps (1%)
[06:46:45] Completed 5000 out of 250000 steps (2%)
[07:56:35] Completed 7500 out of 250000 steps (3%)
[09:04:41] Completed 10000 out of 250000 steps (4%)
[10:21:46] Completed 12500 out of 250000 steps (5%)
Click to expand...


Benchmarks
CPU Blowfish
CPU Blowfish
This Machine 1800 MHz 1.983
Intel(R) Celeron(R) M processor 1.50GHz (null) 26.1876862
PowerPC 740/750 (280.00MHz) (null) 172.816713
CPU CryptoHash
CPU Cryptohash
This Machine 1800 MHz 664.695
CPU Fibonacci
CPU Fibonacci
This Machine 1800 MHz 3.719
Intel(R) Celeron(R) M processor 1.50GHz (null) 8.1375674
PowerPC 740/750 (280.00MHz) (null) 58.07682
CPU N-Queens
CPU N-Queens
This Machine 1800 MHz 1.241
FPU FFT
FPU FFT
This Machine 1800 MHz 3.010
FPU Raytracing
FPU Raytracing
This Machine 1800 MHz 26.069
Intel(R) Celeron(R) M processor 1.50GHz (null) 40.8816714
PowerPC 740/750 (280.00MHz) (null) 161.312647
Click to expand...
 
Hmm. I don't see anything wrong other than the obviously bad TPF :(

You ma need to wait for someone much more knowledgeable than me.
 
ok I received the linux update window (took a lot of time for it to appear). Now I'm scoring 1.5 for the CPU blowfish.

Info for people setting up bigadv under linux:
- Don't go with Ubuntu 8 desktop, it supports maximum 8 cores, go with Ubuntu 9 or else.
- Install 32bit library. For ubuntu/debian: apt-get install ia32-libs
- Wait for the updates before starting
- The tutorial on stanford (http://folding.stanford.edu/English/LinSMPGuide) worked for me.
- You may need to enable sensor in Ubuntuwith "sudo apt-get install lm-sensors". Monitoring tools: GKrellM System Monitor, System Profiler and Benchmark
 
Last edited:
Ohh you where running 8?

9.04 or 9.10 where what I was always told to go with.
 
404 - File not found said:
Yeah, I didn't know about the limitation. I just took a random amd64 release. Never played with linux before..
Click to expand...

Glad you caught that though, because checking the OS version was not on the list of stuff I was thinking could be wrong.

(BTW i didn't see a diff between 9.04 and 9.10)
 
I wouldn't have known about the limitation as my max is still 8 cores. Will change later in the year, but for now, just 8.
 
You must log in or register to reply here.
Back
Top