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

Setting up a new Ubuntu Box

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

chuckerants

Member
Joined
Mar 8, 2002
Location
Phoenix
I have a borrowed Dell Power Edge with a Pentium D Smithfield 3.0 Ghz. I just installed Ubuntu 9.04 x64 and managed to install and run FAH in SMP.

My question is that I see 4 instances of FahCore_a2.exe and the status is "sleeping". The two CPU cores are maxed out according to the System Monitor.

According to FHM and FahMON, I am seeing that the WU is hung.

Does this look right?

Code:
--- Opening Log file [October 22 23:11:10] 


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

                       Folding@Home Client Version 6.02

                          http://folding.stanford.edu

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

Launch directory: /home/chuck/folding
Executable: ./fah6
Arguments: -configonly 

[23:11:10] Configuring Folding@Home...


[23:11:27] - Ask before connecting: No
[23:11:27] - User name: chuckerants (Team 32)
[23:11:27] - User ID not found locally
[23:11:27] + Requesting User ID from server
[23:11:31] - Machine ID: 1
[23:11:31] 
[23:11:31] -configonly flag given, so exiting.


--- Opening Log file [October 22 23:12:18] 


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

                       Folding@Home Client Version 6.02

                          http://folding.stanford.edu

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

Launch directory: /home/chuck/folding
Executable: ./fah6
Arguments: -smp -verbosity 9 

[23:12:18] - Ask before connecting: No
[23:12:18] - User name: chuckerants (Team 32)
[23:12:18] - User ID: 24E0846F24954AC9
[23:12:18] - Machine ID: 1
[23:12:18] 
[23:12:18] Work directory not found. Creating...
[23:12:18] Could not open work queue, generating new queue...
[23:12:18] - Autosending finished units...
[23:12:18] Trying to send all finished work units
[23:12:18] + No unsent completed units remaining.
[23:12:18] - Autosend completed
[23:12:18] - Preparing to get new work unit...
[23:12:18] + Attempting to get work packet
[23:12:18] - Detect CPU. Vendor: GenuineIntel, Family: 15, Model: 4, Stepping: 7
[23:12:18] - Connecting to assignment server
[23:12:18] Connecting to http://assign.stanford.edu:8080/
[23:12:24] Posted data.
[23:12:24] Initial: 40AB; - Successful: assigned to (171.64.65.56).
[23:12:24] + News From Folding@Home: Welcome to Folding@Home
[23:12:24] Loaded queue successfully.
[23:12:24] Connecting to http://171.64.65.56:8080/
[23:12:35] Posted data.
[23:12:35] Initial: 0000; - Receiving payload (expected size: 4836781)
[23:12:40] - Downloaded at ~944 kB/s
[23:12:40] - Averaged speed for that direction ~944 kB/s
[23:12:40] + Received work.
[23:12:40] + Closed connections
[23:12:40] 
[23:12:40] + Processing work unit
[23:12:40] Core required: FahCore_a2.exe
[23:12:40] Core not found.
[23:12:40] - Core is not present or corrupted.
[23:12:40] - Attempting to download new core...
[23:12:40] + Downloading new core: FahCore_a2.exe
[23:12:40] Downloading core (/~pande/Linux/x86/Core_a2.fah from www.stanford.edu)
[23:12:44] Initial: AFDE; + 10240 bytes downloaded
[
[23:12:45] Initial: 0D9E; + 686080 bytes downloaded
SNIPPED
[23:12:46] Initial: D8FF; + 2412195 bytes downloaded
[23:12:46] Verifying core Core_a2.fah...
[23:12:46] Signature is VALID
[23:12:46] 
[23:12:46] Trying to unzip core FahCore_a2.exe
[23:12:47] Decompressed FahCore_a2.exe (5509624 bytes) successfully
[23:12:47] + Core successfully engaged
[23:12:54] 
[23:12:54] + Processing work unit
[23:12:54] Core required: FahCore_a2.exe
[23:12:54] Core found.
[23:12:54] Working on Unit 01 [October 22 23:12:54]
[23:12:54] + Working ...
[23:12:54] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a2.exe -dir work/ -suffix 01 -checkpoint 15 -verbose -lifeline 3282 -version 602'

[23:12:54] 
[23:12:54] *------------------------------*
[23:12:54] Folding@Home Gromacs SMP Core
[23:12:54] Version 2.10 (Sun Aug 30 03:43:28 CEST 2009)
[23:12:54] 
[23:12:54] Preparing to commence simulation
[23:12:54] - Ensuring status. Please wait.
[23:12:56] Called DecompressByteArray: compressed_data_size=4836269 data_size=23979541, decompressed_data_size=23979541 diff=0
[23:12:56] - Digital signature verified
[23:12:56] 
[23:12:56] Project: 2669 (Run 11, Clone 147, Gen 90)
[23:12:56] 
[23:12:56] Assembly optimizations on if available.
[23:12:56] Entering M.D.
[23:13:06] un 11, Clone 147, Gen 90)
[23:13:06] 
[23:13:06] Entering M.D.
 
Here's the Config file:

[settings]
username=chuckerants
team=32
passkey=
asknet=no
machineid=1

[http]
active=no
host=localhost
port=8080

And I set this up just like the VM/Ubuntus using-

./fah6 -smp -verbosity 9
 
Mine goes after that. If its still doesn't show a working status after another 5 minutes, probably stuck. Your proc might just be slow.
 
Gee, I was all excited because HFM was showing 3600 PPD as soon as the WU was received, but it's running at LEAST 20+ minutes per frame instead of the projected 7 minutes per frame.

Oh well, I'll just let this run for a while and see what happens.
 
Wow, could it be this fast???

[23:38:37] Entering M.D.
[23:38:43] Using Gromacs checkpoints
[23:38:52] Resuming from checkpoint
[23:38:52] Verified work/wudata_01.log
[23:38:52] Verified work/wudata_01.trr
[23:38:52] Verified work/wudata_01.xtc
[23:38:52] Verified work/wudata_01.edr
[23:38:53] Completed 1944 out of 250000 steps (0%)
[23:44:24] Completed 2500 out of 250000 steps (1%)

Project ID: 2669
Core: GROCVS
Credit: 1920
Frames: 100


Name: Dell Server
Path: V:\
Number of Frames Observed: 2

Min. Time / Frame : 00:05:31 - 5011.7 PPD
Avg. Time / Frame : 00:06:34 - 4210.4 PPD
Cur. Time / Frame : 00:05:31 - 5011.7 PPD
R3F. Time / Frame : 00:00:00 - 0 PPD
All Time / Frame : 00:05:31 - 5011.7 PPD
Eff. Time / Frame : -06:26:55 - -71.5 PPD
 
It's not that fast. It started from a checkpoint. The new Gromacs checkpoints that came with v2.10 don't loose near as much time as the old FAH checkpoints.
You actually did 22% of the WU in 5:31. You can expect about 25:06/Frame, 1100 ppd.
 
That's better then i expected.

I've found that some SMP units don't give you progress reports, they go at the same speed, but never report the % complete. I've had two or three of 'em.
 
Well, I could dream can't I? :)

Project ID: 2669
Core: GROCVS
Credit: 1920
Frames: 100


Name: Dell Server
Path: V:\
Number of Frames Observed: 6

Min. Time / Frame : 00:05:31 - 5011.7 PPD
Avg. Time / Frame : 00:18:35 - 1487.8 PPD
Cur. Time / Frame : 00:24:36 - 1123.9 PPD
R3F. Time / Frame : 00:24:38 - 1122.4 PPD
All Time / Frame : 00:20:47 - 1330.3 PPD
Eff. Time / Frame : -00:57:01 - -484.9 PPD

Now if only I could figure out why I can't see the Ubuntu share from my W7 desktop.
 
1000 isn't bad considering I'm only getting 3200 from my phenom ii 955 @ 3.8ghz

There's something wrong with that if it's in native ubuntu, the 810 i've been playing with at 3.6 gets 5200-5400, native linux smp4. It isn't being used for anything but folding (or wasn't when it was folding, at least) though.

However, the e1200 (core2duo with tinycache) at 2.6 gets a whopping 1600ppd, the P4D is mighty close to that, really.
 
Back