SETI & SETI BETA

Discussion in 'Distributed Computing' started by Gilthanis, Aug 1, 2014.

  1. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    For those looking to run optimized apps, you can find them here: Seti@Home optimized science apps and information

    We also have a challenge on 8/15/14-8/29/14

    I am adding this app_config info in the first post so that any whom come here will have easy access. This allows for 2 tasks to run on your GPU at a time. You can change the numbers to run more, but it is up to you to decide what maximizes your points on your hardware.:

    <app_config>
    <app>
    <name>astropulse_v7</name>
    <gpu_versions>
    <gpu_usage>0.5</gpu_usage>
    <cpu_usage>0.3</cpu_usage>
    </gpu_versions>
    </app>
    <app>
    <name>setiathome_v7</name>
    <gpu_versions>
    <gpu_usage>0.5</gpu_usage>
    <cpu_usage>0.3</cpu_usage>
    </gpu_versions>
    </app>
    </app_config>
    <app_config>
    <app>
    <name>setiathome_v8</name>
    <gpu_versions>
    <gpu_usage>0.5</gpu_usage>
    <cpu_usage>0.5</cpu_usage>
    </gpu_versions>
    </app>
    </app_config>

    Edit: I created an account called HardOCPtest so that anyone whom wanted to test or contribute anonymously could do so.
    SETI
    &#8226;to attach a computer to your account without using the BOINC Manager. To do so, install BOINC, create a file named account_setiathome.berkeley.edu.xml in the BOINC data directory, and set its contents to:
    <account>
    <master_url>http://setiathome.berkeley.edu/</master_url>
    <authenticator>10047042_37571956872286d7bb338dd0ca7a693b</authenticator>
    </account>

    SETI BETA
    &#8226;to attach a computer to your account without using the BOINC Manager. To do so, install BOINC, create a file named account_setiweb.ssl.berkeley.edu_beta.xml in the BOINC data directory, and set its contents to:
    <account>
    <master_url>http://setiweb.ssl.berkeley.edu/beta/</master_url>
    <authenticator>26285_3b01b7aa8efe59344accd5f76d6e589b</authenticator>
    </account>

    SETI
    [​IMG]

    SETI BETA
    [​IMG]

    Edit again: I edited the app_config above to reflect them moving from v7 to v8. Once v7 is completely finished you can remove it from the app_config
     
    Last edited: Aug 24, 2016
  2. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    Signed for the WOW event and it worked for the first night.
    Since this morning no new assignments come my way. I added GPUGrid to cross check which works ok.

    Any idea why SETI don't give me continuously work ?
    Ubuntu 14.04, CUDA6, two GTX 780 (one for something else in use)

    Since I'm rather new to BOINC (did some few CPU tasks last year and begin of this year) any help and guidance is appreciated!
     
    Last edited: Aug 16, 2014
  3. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Can you take a look at the event log and post the first 30 or so lines? I would recommend doing it from a fresh client start and would also do a manual update of SETI with all other projects suspended so that SETI is the only project requesting work. This will give us clues as to what is going on and your system setup.
     
  4. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    this one is the latest try ... it identify my GPU, and that one is excluded but still not giving any work to the other nV anymore. Funny enough it was ok the first day.

    Code:
    Sun Aug 17 15:17:21 2014 |  | Starting BOINC client version 7.2.42 for x86_64-pc-linux-gnu
    Sun Aug 17 15:17:21 2014 |  | log flags: file_xfer, sched_ops, task
    Sun Aug 17 15:17:21 2014 |  | Libraries: libcurl/7.35.0 OpenSSL/1.0.1f zlib/1.2.8 libidn/1.28 librtmp/2.3
    Sun Aug 17 15:17:21 2014 |  | Data directory: /var/lib/boinc-client
    Sun Aug 17 15:17:21 2014 |  | CUDA: NVIDIA GPU 0: GeForce GTX 780 (driver version unknown, CUDA version 6.0, compute capability 3.5, 3072MB, 2746MB available, 4154 GFLOPS peak)
    Sun Aug 17 15:17:21 2014 |  | CUDA: NVIDIA GPU 1 (ignored by config): GeForce GTX 780 (driver version unknown, CUDA version 6.0, compute capability 3.5, 3072MB, 2910MB available, 4154 GFLOPS peak)
    Sun Aug 17 15:17:21 2014 |  | OpenCL: NVIDIA GPU 0: GeForce GTX 780 (driver version 331.62, device version OpenCL 1.1 CUDA, 3072MB, 2746MB available, 4154 GFLOPS peak)
    Sun Aug 17 15:17:21 2014 |  | OpenCL: NVIDIA GPU 1 (ignored by config): GeForce GTX 780 (driver version 331.62, device version OpenCL 1.1 CUDA, 3072MB, 2910MB available, 4154 GFLOPS peak)
    Sun Aug 17 15:17:21 2014 |  | Host name: linuxpowered
    Sun Aug 17 15:17:21 2014 |  | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2600S CPU @ 2.80GHz [Family 6 Model 42 Stepping 7]
    Sun Aug 17 15:17:21 2014 |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid
    Sun Aug 17 15:17:21 2014 |  | OS: Linux: 3.13.0-29-generic
    Sun Aug 17 15:17:21 2014 |  | Memory: 7.74 GB physical, 27.00 GB virtual
    Sun Aug 17 15:17:21 2014 |  | Disk: 82.38 GB total, 71.10 GB free
    Sun Aug 17 15:17:21 2014 |  | Local time is UTC +9 hours
    Sun Aug 17 15:17:21 2014 |  | Config: ignoring NVIDIA GPU 1
    Sun Aug 17 15:17:21 2014 |  | Config: GUI RPCs allowed from:
    Sun Aug 17 15:17:21 2014 |  | 192.xx.xx.xx
    Sun Aug 17 15:17:21 2014 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7361513; resource share 400
    Sun Aug 17 15:17:21 2014 | SETI@home | General prefs: from SETI@home (last modified 17-Aug-2014 00:10:06)
    Sun Aug 17 15:17:21 2014 | SETI@home | Host location: none
    Sun Aug 17 15:17:21 2014 | SETI@home | General prefs: using your defaults
    Sun Aug 17 15:17:21 2014 |  | Preferences:
    Sun Aug 17 15:17:21 2014 |  | max memory usage when active: 3964.22MB
    Sun Aug 17 15:17:21 2014 |  | max memory usage when idle: 7135.60MB
    Sun Aug 17 15:17:21 2014 |  | max disk usage: 41.19GB
    Sun Aug 17 15:17:21 2014 |  | suspend work if non-BOINC CPU load exceeds 88%
    Sun Aug 17 15:17:21 2014 |  | (to change preferences, visit a project web site or select Preferences in the Manager)
    Sun Aug 17 15:17:21 2014 |  | gui_rpc_auth.cfg is empty - no GUI RPC password protection
    Sun Aug 17 15:17:21 2014 |  | Not using a proxy
    Sun Aug 17 15:20:55 2014 | SETI@home | update requested by user
    Sun Aug 17 15:20:59 2014 | SETI@home | Sending scheduler request: Requested by user.
    Sun Aug 17 15:20:59 2014 | SETI@home | Not requesting tasks: don't need
    Sun Aug 17 15:21:01 2014 | SETI@home | Scheduler request completed
    Sun Aug 17 15:26:07 2014 | SETI@home | Sending scheduler request: To fetch work.
    Sun Aug 17 15:26:07 2014 | SETI@home | Requesting new tasks for NVIDIA
    Sun Aug 17 15:26:09 2014 | SETI@home | Scheduler request completed: got 0 new tasks
    
    are you guys with CPU on the way or also GPU (AMD or nV) ?
     
  5. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    once I allowed usage of CPU too it gave me a bunch of new assignments ... 100 or so ... but still nothing on nV anymore.
     
  6. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Try detaching from all other GPU capable projects. Then see if it will pull work. It could be a resource share issue where it owes time to other projects. Also, double check your preferences on the SETI website to make sure nothing got changed there.
     
  7. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
  8. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    thanks; I dropped in there; let see if I get some info there.

    I stopped all other folding processes; two GTX 780 would be fully for SETI but still I get only CPU

    Code:
    Mon Aug 18 20:22:28 2014 |  | Starting BOINC client version 7.2.42 for x86_64-pc-linux-gnu
    Mon Aug 18 20:22:28 2014 |  | log flags: file_xfer, sched_ops, task
    Mon Aug 18 20:22:28 2014 |  | Libraries: libcurl/7.35.0 OpenSSL/1.0.1f zlib/1.2.8 libidn/1.28 librtmp/2.3
    Mon Aug 18 20:22:28 2014 |  | Data directory: /var/lib/boinc-client
    Mon Aug 18 20:22:28 2014 |  | CUDA: NVIDIA GPU 0: GeForce GTX 780 (driver version unknown, CUDA version 6.0, compute capability 3.5, 3072MB, 2746MB available, 4154 GFLOPS peak)
    Mon Aug 18 20:22:28 2014 |  | CUDA: NVIDIA GPU 1: GeForce GTX 780 (driver version unknown, CUDA version 6.0, compute capability 3.5, 3072MB, 2986MB available, 4154 GFLOPS peak)
    Mon Aug 18 20:22:28 2014 |  | OpenCL: NVIDIA GPU 0: GeForce GTX 780 (driver version 331.62, device version OpenCL 1.1 CUDA, 3072MB, 2746MB available, 4154 GFLOPS peak)
    Mon Aug 18 20:22:28 2014 |  | OpenCL: NVIDIA GPU 1: GeForce GTX 780 (driver version 331.62, device version OpenCL 1.1 CUDA, 3072MB, 2986MB available, 4154 GFLOPS peak)
    Mon Aug 18 20:22:28 2014 |  | Host name: linuxpowered
    Mon Aug 18 20:22:28 2014 |  | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-2600S CPU @ 2.80GHz [Family 6 Model 42 Stepping 7]
    Mon Aug 18 20:22:28 2014 |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid
    Mon Aug 18 20:22:28 2014 |  | OS: Linux: 3.13.0-29-generic
    Mon Aug 18 20:22:28 2014 |  | Memory: 7.74 GB physical, 27.00 GB virtual
    Mon Aug 18 20:22:28 2014 |  | Disk: 82.38 GB total, 71.17 GB free
    Mon Aug 18 20:22:28 2014 |  | Local time is UTC +9 hours
    Mon Aug 18 20:22:28 2014 |  | Config: GUI RPCs allowed from:
    Mon Aug 18 20:22:28 2014 |  | 192.xxx.xxx.xxx
    Mon Aug 18 20:22:28 2014 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7361513; resource share 400
    Mon Aug 18 20:22:28 2014 | SETI@home | General prefs: from SETI@home (last modified 17-Aug-2014 00:10:06)
    Mon Aug 18 20:22:28 2014 | SETI@home | Host location: none
    Mon Aug 18 20:22:28 2014 | SETI@home | General prefs: using your defaults
    Mon Aug 18 20:22:28 2014 |  | Reading preferences override file
    Mon Aug 18 20:22:28 2014 |  | Preferences:
    Mon Aug 18 20:22:28 2014 |  | max memory usage when active: 3964.22MB
    Mon Aug 18 20:22:28 2014 |  | max memory usage when idle: 7135.60MB
    Mon Aug 18 20:22:28 2014 |  | max disk usage: 41.19GB
    Mon Aug 18 20:22:28 2014 |  | suspend work if non-BOINC CPU load exceeds 88%
    Mon Aug 18 20:22:28 2014 |  | (to change preferences, visit a project web site or select Preferences in the Manager)
    Mon Aug 18 20:22:28 2014 |  | gui_rpc_auth.cfg is empty - no GUI RPC password protection
    Mon Aug 18 20:22:28 2014 |  | Not using a proxy
    Mon Aug 18 20:23:56 2014 | SETI@home | update requested by user
    Mon Aug 18 20:24:00 2014 | SETI@home | Sending scheduler request: Requested by user.
    Mon Aug 18 20:24:00 2014 | SETI@home | Not requesting tasks: "no new tasks" requested via Manager
    Mon Aug 18 20:24:04 2014 | SETI@home | Scheduler request completed
    Mon Aug 18 20:24:12 2014 | SETI@home | work fetch resumed by user
    Mon Aug 18 20:24:17 2014 | SETI@home | update requested by user
    Mon Aug 18 20:24:19 2014 | SETI@home | Sending scheduler request: Requested by user.
    Mon Aug 18 20:24:19 2014 | SETI@home | Requesting new tasks for CPU and NVIDIA
    Mon Aug 18 20:24:21 2014 | SETI@home | Scheduler request completed: got 0 new tasks
    Mon Aug 18 20:24:21 2014 | SETI@home | Not sending work - last request too recent: 17 sec
    Mon Aug 18 20:29:28 2014 | SETI@home | Sending scheduler request: To fetch work.
    Mon Aug 18 20:29:28 2014 | SETI@home | Requesting new tasks for CPU and NVIDIA
    Mon Aug 18 20:29:30 2014 | SETI@home | Scheduler request completed: got 45 new tasks
    Mon Aug 18 20:29:32 2014 | SETI@home | Started download of 15se08aa.10825.23385.438086664197.12.189.vlar
    Mon Aug 18 20:29:32 2014 | SETI@home | Started download of 15no08ab.13917.19295.438086664201.12.220
    Mon Aug 18 20:29:35 2014 | SETI@home | Finished download of 15se08aa.10825.23385.438086664197.12.189.vlar
    ...
    Mon Aug 18 20:30:19 2014 | SETI@home | Finished download of 15se08aa.10825.23385.438086664197.12.253.vlar
    Mon Aug 18 20:30:24 2014 | SETI@home | Computation for task 03fe09ae.2433.111158.438086664207.12.122_0 finished
    Mon Aug 18 20:30:24 2014 | SETI@home | Starting task 03fe09ae.2433.111158.438086664207.12.92_1
    Mon Aug 18 20:30:26 2014 | SETI@home | Started upload of 03fe09ae.2433.111158.438086664207.12.122_0_0
    Mon Aug 18 20:30:30 2014 | SETI@home | Finished upload of 03fe09ae.2433.111158.438086664207.12.122_0_0
    
     
  9. RFGuy_KCCO

    RFGuy_KCCO DCOTM x4

    Messages:
    886
    Joined:
    Sep 23, 2006
    Looks like SETI does not have a CUDA app for Linux for MB WU's, only for AP WU's. The problem is there are no AP WU's available at this time. If you were on Windows, you would have both MB and AP WU's available to you, so you could crunch MB WU's when there are no AP WU's available. Looks like this is one of the drawbacks of running Linux, unfortunately.
     
  10. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    Seems to be that my combo is not a good fit with SETI ... I got a small number of opencl-Wu though. But yeah, nothing I can do. Try to get as much as possible done during the WOW event. With not bring us on the podium. :(

    Works great for FAH or GPUGrid ... Back to protein in 10 days
     
  11. RFGuy_KCCO

    RFGuy_KCCO DCOTM x4

    Messages:
    886
    Joined:
    Sep 23, 2006
    Poem is another good medical project you could try. They have an OpenCL app for Nvidia GPU's on Linux. You could easily run 2 WU's at a time on each of your 780's if you use an app_config.xml file, as well. They don't always have GPU work, though.

    If you are interested in space projects, there are also Einstein and Milkyway. I believe both should have Linux apps. Asteroids is another one with Nvidia work.

    If you're interested in cryptography, there is Moo! Wrapper which is trying to crack an RC5-72 encrypted message by trying every key combination through brute force.

    Finally, if you have any interest in mathematics, there is PrimeGrid. Their PPS Sieve project runs like gangbusters on Nvidia GPU's.

    Probably more info than you were looking for, but I wanted to let you know about all the GPU projects you can try out with your Nvidia GPU's.
     
    Last edited: Aug 19, 2014
  12. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    You could also always run SETI BETA or Albert :p Then of course if you don't care about efficiencies, there is BitcoinUtopia.

    However, what I would suggest is to also attach to GPUGrid because they have good bio/medical research and score pretty well. They will occaisionally run out of work but if you are attached to a few projects, you should be fine with staying busy.
     
  13. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    thanks guys for the suggestions; originally goal was to participate in the WOW! event ... but I'm really running dry on WU for a sufficient contribution. I don't have Windows as I prefer Linux for DC-projects. So I just keep my CPU running; whatever it gets. Hopefully there will be a similar challenge on GPUgrid; that worked well for me in the breaks. try Poem again too ...
     
  14. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    hurra, finally SETI send me some ap* package for opencl_nvidia_100

    [H]appy champer here; ok; now I just have to finish some WU from another DC project. But then I try to catch up as long as the WU coming in.
     
  15. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Yes...pretty much every Tuesday is maintenance day. So, if you are wanting to guarantee the work units, make sure to raise your cache...
     
  16. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    One thing I like with SETI: the temps are much lower compared to other DC projects:

    [​IMG]

    Very summer friendly, fan fixed on 80%, GTX 780 for both GPU (red line)
    Blue lines are i7 2600S cores on AIO-WC folding some proteins.
     
  17. Kai

    Kai [H]ard|DCer of the Month - February 2014

    Messages:
    1,000
    Joined:
    May 31, 2005
    How's the core usage though?
     
  18. ChristianVirtual

    ChristianVirtual [H]ard DCOTM x3

    Messages:
    2,522
    Joined:
    Feb 23, 2013
    CPU:6 for FAH, rest for SETI and Ubuntu.

    [​IMG]

    The last minutes since 20:50 GPUGrid is running, temps are higher again.
     
  19. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Yeah... GPUGrid tends to stress cards in ways other projects don't. The demand on the card is different with each batch too. There for a bit I had to underclock a few GPU's because they would error out if I didn't. Since then, they are back to stock settings. Some projects put a priority on not fully utilizing equipment so that it lessens the possibility of reducing the user experience.
     
  20. RFGuy_KCCO

    RFGuy_KCCO DCOTM x4

    Messages:
    886
    Joined:
    Sep 23, 2006
    Neither Seti or Seti Beta have updated any stats since 11/3. I tried poking around the Seti site and message boards, but didn't find anything about this issue. Does anyone know what's going on?
     
  21. jojo69

    jojo69 [H]ardForum Junkie

    Messages:
    10,429
    Joined:
    Sep 13, 2009
    they lost Bruno...one of their main servers, last weekend
     
  22. RFGuy_KCCO

    RFGuy_KCCO DCOTM x4

    Messages:
    886
    Joined:
    Sep 23, 2006
    I'm guessing it has something to do with the post below made by Eric Korpela from Seti, but there is no mention of stats exports:

     
  23. RFGuy_KCCO

    RFGuy_KCCO DCOTM x4

    Messages:
    886
    Joined:
    Sep 23, 2006
    Thanks, jojo69. I kind of figured it had to be related to the Bruno issues.
     
  24. jojo69

    jojo69 [H]ardForum Junkie

    Messages:
    10,429
    Joined:
    Sep 13, 2009
    they failed to even provide work for about 30 hours, at least the heaters are working again
     
  25. jojo69

    jojo69 [H]ardForum Junkie

    Messages:
    10,429
    Joined:
    Sep 13, 2009
    aaaaaaaaand they're gone

    Had to switch over to Rosetta to avoid freezing to death.
     
  26. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Yeah... SETI has lots of server issues. So, make sure to keep a backup project or two if you plan on running them. Also, make sure to run a larger cache. That is what most dedicated SETI DC'ers do.

    However, we are part of a WCG challenge right now. Why not jump over to there and help with a few work units. Just make sure you attach to HardOCP and not [H]ard|OCP. ;)
     
  27. jojo69

    jojo69 [H]ardForum Junkie

    Messages:
    10,429
    Joined:
    Sep 13, 2009
    The new data from Arecibo observatory has been processed and SETI started busting new units about an hour ago, approximately 06:40 UT Saturday.
     
  28. RFGuy_KCCO

    RFGuy_KCCO DCOTM x4

    Messages:
    886
    Joined:
    Sep 23, 2006
    SETI Beta still hasn't exported any stats since 11/3 and a thread about this on their forum hasn't received any response from the admins. I have now removed this project from all of my machines until they can get their act together. I increased my work share at SETI to make up the difference.

    Patience has never been one of my virtues. :rolleyes:
     
  29. 402blownstroker

    402blownstroker [H]ard|DCer of the Month - Nov. 2012

    Messages:
    3,185
    Joined:
    Jan 5, 2006
    Just crossed the 1M credit mark with a RAC of 26,378.32.
     
  30. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Pretty nice achievement for SETI. Those work units can sometimes take months to validate...lol
     
  31. 402blownstroker

    402blownstroker [H]ard|DCer of the Month - Nov. 2012

    Messages:
    3,185
    Joined:
    Jan 5, 2006
    No kidding. I have been running SETI for about a day or two to load up the queue and then switch over to WCG for 3 or 4 days while the SETI queue drains down. Then I repeat the cycle.
     
  32. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    How do you have your cache settings and your priority settings for the projects? Are you just running SETI on GPU or CPU as well?
     
  33. 402blownstroker

    402blownstroker [H]ard|DCer of the Month - Nov. 2012

    Messages:
    3,185
    Joined:
    Jan 5, 2006
    Just running CPUs as there does not appear to be any GPU tasks for linux. I have the cache setup for 2 days + 2 days reserve. My 2 2Ps can burn through that work in about 1.5 days and is about 500 tasks. My 4P burns through that in about the same time frame. After about a day of processing I I set the project not to allow new tasks. When the work queue about matches the number of CPUs/Threads, I update another project to accept new tasks. It is about 6 to 8 hours of overlap between the projects before the other project has all resources. Most likely not the most elegant approach, but it works :)
     
  34. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Well...not knowing why you have your settings the way you do... here are a few suggestions.

    If SETI is your primary project, you can count on their servers being down every week on Monday (I think or it is Tuesday... either way it is weekly same bat time same bat channel). Now, if you are trying to cache enough work, then just keep running a 2-3 day cache. Now, if you are wanting WCG to just be a backup project for when you run out of work, you should just change your project priority level from the default 100 to 0 for WCG. What that does is tell the client to only pull work from it if there is no work available from all other projects. Then when SETI has work again, you will start pulling work from them again. It of course will finish the work in progress according to deadlines and such. You can imagine the time it will save you from having to manually cache up each week and tweak settings.

    However, if y ou are wanting to run WCG more, then it really just depends on how you want it to play out. You can assign your priority levels so that it balances work accordingly. For example if you set SETI to 300 and WCG to 100. SETI will run 3 times as much work (based on wall clock) as WCG. It won't guarantee 3 SETI and 1 WCG running at all times. It just balances the amount of time given to the projects. If for some reason a project is down when it is time for BOINC to call for more work, it will keep chugging away at the other accumulating a "debt". Then when work is available again, it will make up the difference accordingly.

    Let me know if this helps, if you need assistance, or if you had something completely different in mind. :)
     
  35. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
  36. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    Team name has been changed to [H]ard|OCP. You should not have to do anything if you were previously part of HardOCP.
     
  37. jojo69

    jojo69 [H]ardForum Junkie

    Messages:
    10,429
    Joined:
    Sep 13, 2009
    Don't seem to have picked up any actives with the change, still only 5 of us turning in work.

    My output will be much lower for Northern hemisphere summer
     
  38. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    I don't believe there were any actives crunching SETI at the other team. However, this will at least get us a unified team name almost across the board. Only 3 projects left. One of which (WCG) will not be changed for sure as they have refused to assist in this ordeal. The only option we would have there is to abandon the one we use and switch to one we don't. Getting the masses to manually change over is highly unlikely.

    Edit: it also means that if there is a challenge through BOINCStats, I can sign us up for it there...
     
  39. jojo69

    jojo69 [H]ardForum Junkie

    Messages:
    10,429
    Joined:
    Sep 13, 2009
    Thank you for your efforts in this Gilthanis.
     
  40. Gilthanis

    Gilthanis [H]ard|DCer of the Year - 2014

    Messages:
    7,939
    Joined:
    Jan 29, 2006
    I updated the link to the lunatics optimized apps in the first post. Apparently KWSN lost their team founder and owner of the site, so they had to switch domains. There is also a newer release of the optimized apps package than when this thread was started. So, people should look to see if they have the latest or not. This is a good time to start testing if you have any plans in joining in the Wow! event coming up in a month.