7th Annual Holiday Season Race in Folding@Home

Discussion in 'Distributed Computing' started by petrusbroder, Nov 27, 2012.

  1. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    Thanks Peter for the stas !

    On another note, by first bigadv has done 46% in 24 hours (give or take 30 minute), so I should make the deadline ??

    Edit: 50% at 26 hours worst case. Thats 52 for the unit, and I heard that 2.4 days or 57.6 hours is the max. At 52 hours, I make it pretty easy., right ?

    Thats a lot of horsepower. I thought with 24 cores, I would make it easy. Maybe easy with 16 cores @ 2.8 or more on the Intel platform
     
    #226 Markfw900, Dec 19, 2012
    Last edited: Dec 19, 2012
  2. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    In this thread, 10e posted these bios settings for your motherboard:

    http://foldingforum.org/viewtopic.php?f=55&t=16840


    In the BIOS do the following:

    1) Make sure in the memory configuration that "node interleaving" is enabled. It disables NUMA BIOS support and tends to show a very large improvement in folding performance in the order of 25 to 30%
    2) Make sure your maximum P-State is set to 0 (zero)
    3) Disable Cool 'n' Quiet
    4) Disable "PowerNow"


    Looks like I had it backwards for node interleaving. It should be enabled I guess.
     
  3. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    Thanks biodoc, after this unit finises I will look and fix. At 80% now and I think I will make the deadline.
     
  4. petrusbroder

    petrusbroder Lifer

    Joined:
    Nov 28, 2004
    Messages:
    12,610
    Likes Received:
    5
    The last day of the PrimeGrid Race! Tomorrow I'll rebalance the teams!

    [​IMG]
     
  5. GLeeM

    GLeeM Elite Member

    Joined:
    Apr 2, 2004
    Messages:
    6,932
    Likes Received:
    2
    Thanks for the stats, Peter :)
     
  6. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Yep, thanks Peter! :)
     
  7. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    OK, now I have a critical problem. Its been at 99,99% for 20 minutes, but the ETA said 23 seconds, something is wrong. What do I do ?
    Log file below:
    OK, I fixed the HFM problem, but it won't finish the 8101 that it spent 2 days running, and if I don't get it working soon, I wil loose the 200k bonus. WTF ??? Log file below:

    *********************** Log Started 2012-12-21T03:01:25Z ***********************
    03:01:25:************************* Folding@home Client *************************
    03:01:25: Website: http://folding.stanford.edu/
    03:01:25: Copyright: (c) 2009-2012 Stanford University
    03:01:25: Author: Joseph Coffland <joseph@cauldrondevelopment.com>
    03:01:25: Args: --child --lifeline 1125 /etc/fahclient/config.xml --run-as
    03:01:25: fahclient --pid-file=/var/run/fahclient.pid --daemon
    03:01:25: Config: /etc/fahclient/config.xml
    03:01:25:******************************** Build ********************************
    03:01:25: Version: 7.2.9
    03:01:25: Date: Oct 3 2012
    03:01:25: Time: 10:34:06
    03:01:25: SVN Rev: 3578
    03:01:25: Branch: fah/trunk/client
    03:01:25: Compiler: GNU 4.4.7
    03:01:25: Options: -std=gnu++98 -O3 -funroll-loops -mfpmath=sse -ffast-math
    03:01:25: -fno-unsafe-math-optimizations -msse2
    03:01:25: Platform: linux2 3.2.0-1-amd64
    03:01:25: Bits: 64
    03:01:25: Mode: Release
    03:01:25:******************************* System ********************************
    03:01:25: CPU: AMD Opteron(TM) Processor 6234
    03:01:25: CPU ID: AuthenticAMD Family 21 Model 1 Stepping 2
    03:01:25: CPUs: 24
    03:01:25: Memory: 31.36GiB
    03:01:25:Free Memory: 30.76GiB
    03:01:25: Threads: POSIX_THREADS
    03:01:25: On Battery: false
    03:01:25: UTC offset: -8
    03:01:25: PID: 1206
    03:01:25: CWD: /var/lib/fahclient
    03:01:25: OS: Linux 3.2.0-35-generic x86_64
    03:01:25: OS Arch: AMD64
    03:01:25: GPUs: 2
    03:01:25: GPU 0: NVIDIA:2 GF104 [GeForce GTX 460]
    03:01:25: GPU 1: NVIDIA:2 GF104 [GeForce GTX 460]
    03:01:25: CUDA: 2.1
    03:01:25:CUDA Driver: 4020
    03:01:25:***********************************************************************
    03:01:25:<config>
    03:01:25: <!-- Folding Slot Configuration -->
    03:01:25: <gpu v='true'/>
    03:01:25:
    03:01:25: <!-- Network -->
    03:01:25: <proxy v=':8080'/>
    03:01:25:
    03:01:25: <!-- Remote Command Server -->
    03:01:25: <command-allow v='127.0.0.1,192.168.1.2-192.168.1.255'/>
    03:01:25: <command-allow-no-pass v='127.0.0.1,192.168.1.2-192.168.1.255'/>
    03:01:25:
    03:01:25: <!-- User Information -->
    03:01:25: <passkey v='********************************'/>
    03:01:25: <team v='198'/>
    03:01:25: <user v='Mark_F_Williams'/>
    03:01:25:
    03:01:25: <!-- Folding Slots -->
    03:01:25: <slot id='0' type='GPU'>
    03:01:25: <client-type v='beta'/>
    03:01:25: </slot>
    03:01:25: <slot id='1' type='GPU'>
    03:01:25: <client-type v='beta'/>
    03:01:25: </slot>
    03:01:25: <slot id='2' type='SMP'>
    03:01:25: <client-type v='bigadv'/>
    03:01:25: <cpus v='-1'/>
    03:01:25: <max-packet-size v='big'/>
    03:01:25: </slot>
    03:01:25:</config>
    03:01:25:Switching to user fahclient
    03:01:25:Connecting to assign-GPU.stanford.edu:32762
    \x1b[93m03:01:25:WARNING:Attempting to update GPUs.txt: Could not get IP address for assign-GPU.stanford.edu: No address associated with hostname\x1b[0m
    03:01:25:Trying to access database...
    03:01:25:Successfully acquired database lock
    03:01:25:Enabled folding slot 00: READY gpu:0:\"GF104 [GeForce GTX 460]\"
    03:01:25:Enabled folding slot 01: READY gpu:1:\"GF104 [GeForce GTX 460]\"
    03:01:25:Enabled folding slot 02: READY smp:24
    03:01:25:WU03:FS02:Starting
    03:01:25:WU03:FS02:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/www.stanford.edu/~pande/Linux/AMD64/Core_a5.fah/FahCore_a5 -dir 03 -suffix 01 -version 702 -lifeline 1206 -checkpoint 15 -np 24
    03:01:25:WU03:FS02:Started FahCore on PID 1241
    03:01:25:WU03:FS02:Core PID:1247
    03:01:25:WU03:FS02:FahCore 0xa5 started
    03:01:25:WU00:FS00:Connecting to assign-GPU.stanford.edu:80
    03:01:25:WU01:FS01:Connecting to assign-GPU.stanford.edu:80
    \x1b[93m03:01:25:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:80': Failed to connect to assign-GPU.stanford.edu:80: Network is unreachable\x1b[0m
    03:01:25:WU00:FS00:Connecting to assign-GPU.stanford.edu:8080
    \x1b[93m03:01:25:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Failed to connect to assign-GPU.stanford.edu:80: Network is unreachable\x1b[0m
    03:01:25:WU01:FS01:Connecting to assign-GPU.stanford.edu:8080
    \x1b[93m03:01:25:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Failed to connect to assign-GPU.stanford.edu:8080: Network is unreachable\x1b[0m
    \x1b[91m03:01:25:ERROR:WU00:FS00:Exception: Could not get an assignment\x1b[0m
    \x1b[93m03:01:25:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Failed to connect to assign-GPU.stanford.edu:8080: Network is unreachable\x1b[0m
    \x1b[91m03:01:25:ERROR:WU01:FS01:Exception: Could not get an assignment\x1b[0m
    03:01:26:WU00:FS00:Connecting to assign-GPU.stanford.edu:80
    \x1b[93m03:01:26:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:80': Failed to connect to assign-GPU.stanford.edu:80: Network is unreachable\x1b[0m
    03:01:26:WU00:FS00:Connecting to assign-GPU.stanford.edu:8080
    \x1b[93m03:01:26:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Failed to connect to assign-GPU.stanford.edu:8080: Network is unreachable\x1b[0m
    \x1b[91m03:01:26:ERROR:WU00:FS00:Exception: Could not get an assignment\x1b[0m
    03:01:26:WU01:FS01:Connecting to assign-GPU.stanford.edu:80
    \x1b[93m03:01:26:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Failed to connect to assign-GPU.stanford.edu:80: Network is unreachable\x1b[0m
    03:01:26:WU01:FS01:Connecting to assign-GPU.stanford.edu:8080
    \x1b[93m03:01:26:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Failed to connect to assign-GPU.stanford.edu:8080: Network is unreachable\x1b[0m
    \x1b[91m03:01:26:ERROR:WU01:FS01:Exception: Could not get an assignment\x1b[0m
    03:01:26:WU03:FS02:0xa5:
    03:01:26:WU03:FS02:0xa5:*------------------------------*
    03:01:26:WU03:FS02:0xa5:Folding@Home Gromacs SMP Core
    03:01:26:WU03:FS02:0xa5:Version 2.27 (Thu Feb 10 09:46:40 PST 2011)
    03:01:26:WU03:FS02:0xa5:
    03:01:26:WU03:FS02:0xa5:preparing to commence simulation
    03:01:26:WU03:FS02:0xa5:- Ensuring status. Please wait.
    03:01:29:WU03:FS02:0xa5:- Looking at optimizations...
    03:01:29:WU03:FS02:0xa5:- Working with standard loops on this execution.
    03:01:58:Server connection id=1 on 0.0.0.0:36330 from 127.0.0.1
    03:02:26:WU00:FS00:Connecting to assign-GPU.stanford.edu:80
    03:02:26:WU00:FS00:News: Welcome to Folding@Home
    \x1b[93m03:02:26:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment\x1b[0m
    03:02:26:WU00:FS00:Connecting to assign-GPU.stanford.edu:8080
    03:02:26:WU01:FS01:Connecting to assign-GPU.stanford.edu:80
    03:02:26:WU00:FS00:News: Welcome to Folding@Home
    \x1b[93m03:02:26:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment\x1b[0m
    \x1b[91m03:02:26:ERROR:WU00:FS00:Exception: Could not get an assignment\x1b[0m
    03:02:26:WU01:FS01:News: Welcome to Folding@Home
    \x1b[93m03:02:26:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment\x1b[0m
    03:02:26:WU01:FS01:Connecting to assign-GPU.stanford.edu:8080
    03:02:26:WU01:FS01:News: Welcome to Folding@Home
    \x1b[93m03:02:26:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment\x1b[0m
    \x1b[91m03:02:26:ERROR:WU01:FS01:Exception: Could not get an assignment\x1b[0m
    03:02:26:Server connection id=2 on 0.0.0.0:36330 from 192.168.1.2
    03:04:03:WU00:FS00:Connecting to assign-GPU.stanford.edu:80
    03:04:03:WU00:FS00:News: Welcome to Folding@Home
    \x1b[93m03:04:03:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment\x1b[0m
    03:04:03:WU00:FS00:Connecting to assign-GPU.stanford.edu:8080
    03:04:03:WU00:FS00:News: Welcome to Folding@Home
    \x1b[93m03:04:03:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment\x1b[0m
    \x1b[91m03:04:03:ERROR:WU00:FS00:Exception: Could not get an assignment\x1b[0m
    03:04:03:WU01:FS01:Connecting to assign-GPU.stanford.edu:80
    03:04:03:WU01:FS01:News: Welcome to Folding@Home
    \x1b[93m03:04:03:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment\x1b[0m
    03:04:03:WU01:FS01:Connecting to assign-GPU.stanford.edu:8080
    03:04:03:WU01:FS01:News: Welcome to Folding@Home
    \x1b[93m03:04:03:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment\x1b[0m
    \x1b[91m03:04:03:ERROR:WU01:FS01:Exception: Could not get an assignment\x1b[0m
    03:06:40:WU00:FS00:Connecting to assign-GPU.stanford.edu:80
    03:06:40:WU00:FS00:News: Welcome to Folding@Home
    \x1b[93m03:06:40:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment\x1b[0m
    03:06:40:WU00:FS00:Connecting to assign-GPU.stanford.edu:8080
    03:06:40:WU00:FS00:News: Welcome to Folding@Home
    \x1b[93m03:06:40:WARNING:WU00:FS00:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment\x1b[0m
    \x1b[91m03:06:40:ERROR:WU00:FS00:Exception: Could not get an assignment\x1b[0m
    03:06:40:WU01:FS01:Connecting to assign-GPU.stanford.edu:80
    03:06:40:WU01:FS01:News: Welcome to Folding@Home
    \x1b[93m03:06:40:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment\x1b[0m
    03:06:40:WU01:FS01:Connecting to assign-GPU.stanford.edu:8080
    03:06:41:WU01:FS01:News: Welcome to Folding@Home
    \x1b[93m03:06:41:WARNING:WU01:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment\x1b[0m
    \x1b[91m03:06:41:ERROR:WU01:FS01:Exception: Could not get an assignment\x1b[0m
     
    #232 Markfw900, Dec 20, 2012
    Last edited: Dec 20, 2012
  8. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    OK, it finally finished, I hope I get credit
     
  9. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Jeez, your log file is littered with requests for gpu WU assignments. FAH doesn't support GPUs on linux! How do you find out if you got the bonus? I don't see any way to track specific WUs.
     
  10. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    I removed them and rebooted, but I did not get credit, at least not with the bonus, but I made it in less than 2.4 days. If I don't get credit for the next one, I am quitting F@H. I have used this passkey for years, and turned in bigadv units with it when only 8 threads were required for years.
     
  11. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Did you make the changes in the bios? Hopefully that will make a difference.
     
  12. Uppsala9496

    Uppsala9496 Diamond Member

    Joined:
    Nov 2, 2001
    Messages:
    5,187
    Likes Received:
    2
    Run my guide if you want to use the gpus.
     
  13. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    MarkFW, I just took a look at your thread over at the folding forum. They posted the solution to the very long write times with ext4 filesystems at the end of bigadv WUs. Here's the solution instruction thread. http://hardforum.com/showthread.php?t=1695762


    paste this into a terminal window and hit return:

    sudo sed -ri '/barrier=/ {p;d}; /ext[34]/ {s/([ \t]ext[34][ \t]+)([^ \t]+)([ \t])/\1\2,barrier=0\3/}' /etc/fstab

    then:

    cat /etc/fstab

    You should see something like this:

    UUID=52813dfc-e0e6-461b-9652-49c2ccff0077 / ext3 errors=remount-ro,barrier=0 0 1

    then reboot.

    You are almost there, I think! :thumbsup:
     
  14. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    Yes, I could only find 2 of the 4 things to change. It looks like its doing 31 minutes a frame.

    but I have 2 more problems.

    edit: I saw above, and it worked. Now to the downloading problem....

    Second, it keep downloading a unit in advance, so I will never get a bonus. How do I stop that in v7 client ? I can't find a setting for that.
     
    #239 Markfw900, Dec 21, 2012
    Last edited: Dec 21, 2012
  15. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    I thought downloading the next wu in advance was a good thing. It affects the bonus? I'll do some searching. Wait, now I understand.
     
  16. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Ok, found the solution. You know where you added the extra slot option, like client-type, bigadv? Add another:

    Name: next-unit-percentage
    Value: 100

    This setting will download the next wu when you hit 100%.
     
  17. petrusbroder

    petrusbroder Lifer

    Joined:
    Nov 28, 2004
    Messages:
    12,610
    Likes Received:
    5
    Here are the Friday's stats:

    [​IMG]

    Today the PrimeGrid race ended. I have rebalanced the teams.
    This is how it looks like now:

    [​IMG][​IMG]

    If anyone has some serious objections, please post here.
    But - as we have done in the previous races - now the real race starts. The teams will not change after tomorrows stats. Fold on, TeAm!
     
    #242 petrusbroder, Dec 21, 2012
    Last edited: Dec 21, 2012
  18. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Ok with me! :)

    Thanks for the stats, Peter! :thumbsup:
     
  19. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    OK, all problems resolved, and 30 minutes TPF for 100k ppd !
     
  20. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Awesome!!:cool:
     
  21. Uppsala9496

    Uppsala9496 Diamond Member

    Joined:
    Nov 2, 2001
    Messages:
    5,187
    Likes Received:
    2
    Peter thanks for the stats. Best of luck to the racers!
     
  22. biodoc

    biodoc Diamond Member

    Joined:
    Dec 29, 2005
    Messages:
    4,602
    Likes Received:
    1
    Hopefully the Firecats are holding back some hardware.

    If not, it won't be much of a race! :twisted:

    Hey, someone has to start the trash talk! :biggrin::whiste:
     
  23. Pokey

    Pokey Platinum Member

    Joined:
    Oct 20, 1999
    Messages:
    2,476
    Likes Received:
    6
    Well, with Mark’s big rig now in the race you might be right; maybe santa will bring the Firecats some additional horsepower……….….. Santa?? You there? Hello? ():)
     
  24. Markfw900

    Markfw900 Super Moderator <br>Elite Member<br> CPUs & Overcl
    Super Moderator

    Joined:
    May 16, 2002
    Messages:
    13,486
    Likes Received:
    26
    I have another 100k in power, holding back since its too warm here, If we get below freezing, I might be able to fire one of them up. (and if it doesn't blow the breakers.)
     
  25. PCTC2

    PCTC2 Diamond Member

    Joined:
    Feb 18, 2007
    Messages:
    3,865
    Likes Received:
    1
    Still kind of weird that I don't get bigadv units on my 32 thread boxes... I've been crunching on this passkey since that last time I raced a few years ago.
     
Loading...