configuration for cpu only wu?

Forums › Cruncher's Corner

alephnull
alephnull
Joined: 2 Dec 05
Posts: 21
Credit: 28551947
RAC: 0

RE: Oh, and piece of

23 Mar 2011 3:28:43 UTC
Message 104761 in response to message 104760
(moderation:
    )

    Quote:

    Oh, and piece of advice, never post your real on any forum. It's your unique key to log in with (on this project).


    thanks for that, one day i would probably have not considered this and made this mistake. good to know.

    yup, all seems well now with the work units. again, thanks for everything.

    rob

    alephnull
    alephnull
    Joined: 2 Dec 05
    Posts: 21
    Credit: 28551947
    RAC: 0

    well, things looked good last

    23 Mar 2011 13:12:36 UTC
    Message 104762 in response to message 104761
    (moderation:
      )

      well, things looked good last night i thought then this morning:

      689 Einstein@Home 2011-03-23 06:27:49 Sending scheduler request: To fetch work.
      690 Einstein@Home 2011-03-23 06:27:49 Requesting new tasks for GPU
      695 Einstein@Home 2011-03-23 06:27:50 Scheduler request completed: got 1 new tasks
      ...
      718 Einstein@Home 2011-03-23 06:52:37 Requesting new tasks for GPU
      719 Einstein@Home 2011-03-23 06:52:40 Scheduler request completed: got 1 new tasks
      ...
      729 Einstein@Home 2011-03-23 06:59:47 Requesting new tasks for GPU
      730 Einstein@Home 2011-03-23 06:59:50 Scheduler request completed: got 1 new tasks
      ...
      760 Einstein@Home 2011-03-23 07:10:16 Requesting new tasks for GPU
      767 Einstein@Home 2011-03-23 07:10:20 Scheduler request completed: got 1 new tasks
      ...
      779 Einstein@Home 2011-03-23 07:11:22 Reporting 1 completed tasks, requesting new tasks for GPU
      780 Einstein@Home 2011-03-23 07:11:25 Scheduler request completed: got 1 new tasks
      ...
      839 Einstein@Home 2011-03-23 08:33:18 Sending scheduler request: To fetch work.
      840 Einstein@Home 2011-03-23 08:33:18 Requesting new tasks for GPU
      841 Einstein@Home 2011-03-23 08:33:21 Scheduler request completed: got 1 new tasks
      ...
      854 Einstein@Home 2011-03-23 08:35:09 Requesting new tasks for GPU
      861 Einstein@Home 2011-03-23 08:35:12 Scheduler request completed: got 1 new tasks
      ...
      895 Einstein@Home 2011-03-23 08:54:41 Requesting new tasks for GPU
      896 Einstein@Home 2011-03-23 08:54:43 Scheduler request completed: got 1 new tasks

      rechecked all the settings and they are still all the same. in all, it currently has 17 einstein cuda wu. the good news is that the 570s really take care of the wu in short order.

      im thinking maybe creating an app_info.xml file now to restrict work to cpu or is this not thinking clearly? dont really know what else can be done besides that maybe.

      any thoughts?

      Jord
      Joined: 26 Jan 05
      Posts: 2952
      Credit: 5893653
      RAC: 0

      RE: im thinking maybe

      23 Mar 2011 21:12:54 UTC
      Message 104763 in response to message 104762
      (moderation:
        )

        Quote:

        im thinking maybe creating an app_info.xml file now to restrict work to cpu or is this not thinking clearly? dont really know what else can be done besides that maybe.

        any thoughts?


        Well, if it's a bug at Einstein's servers, then you adding an anonymous platform file isn't going to help much when it comes to fixing the bug. I can only point out this thread to Bernd and let him check whether or not this is something broken on their end or not.

        alephnull
        alephnull
        Joined: 2 Dec 05
        Posts: 21
        Credit: 28551947
        RAC: 0

        RE: RE: im thinking maybe

        23 Mar 2011 21:22:25 UTC
        Message 104764 in response to message 104763
        (moderation:
          )

          Quote:
          Quote:

          im thinking maybe creating an app_info.xml file now to restrict work to cpu or is this not thinking clearly? dont really know what else can be done besides that maybe.

          any thoughts?


          Well, if it's a bug at Einstein's servers, then you adding an anonymous platform file isn't going to help much when it comes to fixing the bug. I can only point out this thread to Bernd and let him check whether or not this is something broken on their end or not.


          gotcha. if you think its better, i can wait and see what the results of those findings are during which time ill set this host back to WCG/CPDN and use a different host for cpu/gpu einstein tasks. eventually it would be cool to dedicate this host to einstein cpu tasks if possible.

          thanks again for the information.

          rob

          Jord
          Joined: 26 Jan 05
          Posts: 2952
          Credit: 5893653
          RAC: 0

          RE: eventually it would be

          23 Mar 2011 22:01:07 UTC
          Message 104765 in response to message 104764
          (moderation:
            )

            Quote:
            eventually it would be cool to dedicate this host to einstein cpu tasks if possible.


            And what are you going to do with those 2 GTX 570s then?
            Play Crysis 2? :P

            alephnull
            alephnull
            Joined: 2 Dec 05
            Posts: 21
            Credit: 28551947
            RAC: 0

            RE: RE: eventually it

            23 Mar 2011 23:48:11 UTC
            Message 104766 in response to message 104765
            (moderation:
              )

              Quote:
              Quote:
              eventually it would be cool to dedicate this host to einstein cpu tasks if possible.

              And what are you going to do with those 2 GTX 570s then?
              Play Crysis 2? :P


              this particular machine has been set up to crunch gpugrid on the 570s hence the reason to try to set up cpu only tasks for einstein.

              i stopped gaming several years ago when the ee servers were taken offline. i dabble in wow an starcraft2 occasionally but all my hardware and maintenance is now dedicated to various boinc projects. as of now most hardware here will support these newer games but there doesnt seem to be any time for that anymore. im sure a lot of people out there would be upset to see good hardware going to waste as it is but better to use it to crunch than let it all go idle i suppose.

              even my ps3 only does folding@home now. cant remember the last time i actually used it for anything else.

              till next time.

              Jord
              Joined: 26 Jan 05
              Posts: 2952
              Credit: 5893653
              RAC: 0

              Could you do us (the

              24 Mar 2011 12:23:24 UTC
              Message 104767 in response to message 104766
              (moderation:
                )

                Could you do us (the investigators) a favor and let that computer continue to fetch work from Einstein? That way we can check whether this is the bug where GPU work is sent even on CPU requests for work, or not.

                With thanks,

                alephnull
                alephnull
                Joined: 2 Dec 05
                Posts: 21
                Credit: 28551947
                RAC: 0

                RE: Could you do us (the

                24 Mar 2011 21:15:37 UTC
                Message 104768 in response to message 104767
                (moderation:
                  )

                  Quote:

                  Could you do us (the investigators) a favor and let that computer continue to fetch work from Einstein? That way we can check whether this is the bug where GPU work is sent even on CPU requests for work, or not.

                  With thanks,


                  ok. that machine is now allowed new work for einstein. i did not change any of the settings. some messages after this action:

                  814 Cosmology@Home 2011-03-24 17:11:32 work fetch suspended by user
                  815 Einstein@Home 2011-03-24 17:11:35 work fetch resumed by user
                  816 Einstein@Home 2011-03-24 17:11:36 Sending scheduler request: To fetch work.
                  817 Einstein@Home 2011-03-24 17:11:36 Requesting new tasks for CPU
                  818 Einstein@Home 2011-03-24 17:11:38 Scheduler request completed: got 5 new tasks
                  819 Einstein@Home 2011-03-24 17:11:40 Started download of PM0068_03451_12.binary
                  820 Einstein@Home 2011-03-24 17:11:40 Started download of PM0068_03451_13.binary
                  821 Einstein@Home 2011-03-24 17:11:54 Finished download of PM0068_03451_12.binary
                  822 Einstein@Home 2011-03-24 17:11:54 Started download of PM0068_03451_14.binary
                  823 Einstein@Home 2011-03-24 17:11:55 Finished download of PM0068_03451_13.binary
                  824 Einstein@Home 2011-03-24 17:11:55 Started download of PM0068_03451_15.binary
                  825 Einstein@Home 2011-03-24 17:12:04 Finished download of PM0068_03451_14.binary
                  826 Einstein@Home 2011-03-24 17:12:04 Finished download of PM0068_03451_15.binary
                  827 Einstein@Home 2011-03-24 17:12:04 Started download of PM0068_03441_384.binary
                  828 Einstein@Home 2011-03-24 17:12:04 Started download of PM0068_03441_385.binary
                  829 Einstein@Home 2011-03-24 17:12:14 Finished download of PM0068_03441_385.binary
                  830 Einstein@Home 2011-03-24 17:12:14 Started download of PM0068_03441_386.binary
                  831 Einstein@Home 2011-03-24 17:12:16 Finished download of PM0068_03441_384.binary
                  832 Einstein@Home 2011-03-24 17:12:16 Started download of PM0068_03441_387.binary
                  833 Einstein@Home 2011-03-24 17:12:23 Finished download of PM0068_03441_386.binary
                  834 Einstein@Home 2011-03-24 17:12:23 Started download of PM0057_027B1_44.binary
                  835 Einstein@Home 2011-03-24 17:12:27 Finished download of PM0068_03441_387.binary
                  836 Einstein@Home 2011-03-24 17:12:27 Started download of PM0057_027B1_45.binary
                  837 Einstein@Home 2011-03-24 17:12:35 Finished download of PM0057_027B1_44.binary
                  838 Einstein@Home 2011-03-24 17:12:35 Started download of PM0057_027B1_46.binary
                  839 Einstein@Home 2011-03-24 17:12:38 Finished download of PM0057_027B1_45.binary
                  840 Einstein@Home 2011-03-24 17:12:38 Started download of PM0057_027B1_47.binary
                  841 Einstein@Home 2011-03-24 17:12:47 Finished download of PM0057_027B1_46.binary
                  842 Einstein@Home 2011-03-24 17:12:47 Started download of PM0068_03441_420.binary
                  843 Einstein@Home 2011-03-24 17:12:50 Finished download of PM0057_027B1_47.binary
                  844 Einstein@Home 2011-03-24 17:12:50 Started download of PM0068_03441_421.binary
                  845 Einstein@Home 2011-03-24 17:12:58 Finished download of PM0068_03441_420.binary
                  846 Einstein@Home 2011-03-24 17:12:58 Started download of PM0068_03441_422.binary
                  847 Einstein@Home 2011-03-24 17:13:02 Finished download of PM0068_03441_421.binary
                  848 Einstein@Home 2011-03-24 17:13:02 Started download of PM0068_03441_423.binary
                  849 Einstein@Home 2011-03-24 17:13:11 Finished download of PM0068_03441_422.binary
                  850 Einstein@Home 2011-03-24 17:13:11 Started download of PM0068_03441_388.binary
                  851 Einstein@Home 2011-03-24 17:13:15 Finished download of PM0068_03441_423.binary
                  852 Einstein@Home 2011-03-24 17:13:15 Started download of PM0068_03441_389.binary
                  853 Einstein@Home 2011-03-24 17:13:24 Finished download of PM0068_03441_388.binary
                  854 Einstein@Home 2011-03-24 17:13:24 Started download of PM0068_03441_390.binary
                  855 Einstein@Home 2011-03-24 17:13:28 Finished download of PM0068_03441_389.binary
                  856 Einstein@Home 2011-03-24 17:13:28 Started download of PM0068_03441_391.binary
                  857 Einstein@Home 2011-03-24 17:13:37 Finished download of PM0068_03441_390.binary
                  858 Einstein@Home 2011-03-24 17:13:41 Finished download of PM0068_03441_391.binary

                  that looks good huh? let me know if theres anything else you would like me to do.

                  Richard Haselgrove
                  Richard Haselgrove
                  Joined: 10 Dec 05
                  Posts: 2143
                  Credit: 3054963251
                  RAC: 2155247

                  Preserving the matching

                  24 Mar 2011 21:52:05 UTC
                  Message 104769 in response to message 104768
                  (moderation:
                    )

                    Preserving the matching server log for analysis (though it looks like everything was assigned to CPU this time as per preferences - can you confirm that at your end, alephnull?):

                    2011-03-24 21:11:38.5034 [PID=9363] Request: [USER#xxxxx] [HOST#2335620] [IP xxx.xxx.xxx.98] client 6.10.58
                    2011-03-24 21:11:38.5125 [PID=9363 ] [send] effective_ncpus 6 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] effective_ngpus 2 max_jobs_on_host_gpu 999999
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] Not using matchmaker scheduling; Not using EDF sim
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] CPU: req 1.00 sec, 5.00 instances; est delay 0.00
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] CUDA: req 0.00 sec, 0.00 instances; est delay 0.00
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] work_req_seconds: 1.00 secs
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] available disk 78.22 GB, work_buf_min 0
                    2011-03-24 21:11:38.5126 [PID=9363 ] [send] active_frac 0.999527 on_frac 0.974845 DCF 1.554205
                    2011-03-24 21:11:38.5135 [PID=9363 ] [send] [HOST#2335620] is reliable
                    2011-03-24 21:11:38.5135 [PID=9363 ] [send] set_trust: random choice for error rate 0.000010: yes
                    2011-03-24 21:11:38.5195 [PID=9363 ] [version] Checking plan class 'BRP3SSE'
                    2011-03-24 21:11:38.5196 [PID=9363 ] [version] Checking plan class 'BRP3cuda32'
                    2011-03-24 21:11:38.5196 [PID=9363 ] [version] CUDA app estimated 33.02 GFLOPS (clock 1594000 count 15): using plan class (BRP3cuda32)
                    2011-03-24 21:11:38.5196 [PID=9363 ] [version] Don't need CUDA jobs, skipping version 107 for einsteinbinary_BRP3 (BRP3cuda32)
                    2011-03-24 21:11:38.5196 [PID=9363 ] [version] Best version of app einsteinbinary_BRP3 is ID 256 (3.30 GFLOPS)
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] Sorted list of URLs follows [host timezone: UTC-14400]
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] zone=-21600 url=http://einstein-dl3.phys.uwm.edu
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] zone=-21600 url=http://einstein-dl2.phys.uwm.edu
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] zone=-21600 url=http://einstein-dl4.phys.uwm.edu
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] zone=-28800 url=http://einstein.ligo.caltech.edu
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] zone=+03600 url=http://einstein-mirror.aei.uni-hannover.de/EatH
                    2011-03-24 21:11:38.5211 [PID=9363 ] [debug] zone=+03600 url=http://einstein.aei.mpg.de
                    2011-03-24 21:11:38.5213 [PID=9363 ] [send] [HOST#2335620] Sending app_version einsteinbinary_BRP3 2 105 BRP3SSE; 3.30 GFLOPS
                    2011-03-24 21:11:38.5230 [PID=9363 ] [send] est. duration for WU 95079202: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5230 [PID=9363 ] [HOST#2335620] Sending [RESULT#224692827 PM0068_03451.dm_12_0] (est. dur. 67631.24 seconds)
                    2011-03-24 21:11:38.5237 [PID=9363 ] [send] est. duration for WU 95079035: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5238 [PID=9363 ] [send] [WU#95079035] meets deadline: 67631.24 + 67631.24 < 1209600
                    2011-03-24 21:11:38.5244 [PID=9363 ] [send] [HOST#2335620] Sending app_version einsteinbinary_BRP3 2 105 BRP3SSE; 3.30 GFLOPS
                    2011-03-24 21:11:38.5256 [PID=9363 ] [send] est. duration for WU 95079035: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5256 [PID=9363 ] [HOST#2335620] Sending [RESULT#224692335 PM0068_03441.dm_384_0] (est. dur. 67631.24 seconds)
                    2011-03-24 21:11:38.5265 [PID=9363 ] [send] est. duration for WU 94291468: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5265 [PID=9363 ] [send] [WU#94291468] meets deadline: 135262.49 + 67631.24 < 1209600
                    2011-03-24 21:11:38.5272 [PID=9363 ] [send] [HOST#2335620] Sending app_version einsteinbinary_BRP3 2 105 BRP3SSE; 3.30 GFLOPS
                    2011-03-24 21:11:38.5286 [PID=9363 ] [send] est. duration for WU 94291468: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5286 [PID=9363 ] [HOST#2335620] Sending [RESULT#224692759 PM0057_027B1.dm_44_2] (est. dur. 67631.24 seconds)
                    2011-03-24 21:11:38.5300 [PID=9363 ] [send] est. duration for WU 95079045: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5300 [PID=9363 ] [send] [WU#95079045] meets deadline: 202893.73 + 67631.24 < 1209600
                    2011-03-24 21:11:38.5307 [PID=9363 ] [send] [HOST#2335620] Sending app_version einsteinbinary_BRP3 2 105 BRP3SSE; 3.30 GFLOPS
                    2011-03-24 21:11:38.5318 [PID=9363 ] [send] est. duration for WU 95079045: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5318 [PID=9363 ] [HOST#2335620] Sending [RESULT#224692359 PM0068_03441.dm_420_1] (est. dur. 67631.24 seconds)
                    2011-03-24 21:11:38.5324 [PID=9363 ] [send] est. duration for WU 95079037: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5324 [PID=9363 ] [send] [WU#95079037] meets deadline: 270524.97 + 67631.24 < 1209600
                    2011-03-24 21:11:38.5331 [PID=9363 ] [send] [HOST#2335620] Sending app_version einsteinbinary_BRP3 2 105 BRP3SSE; 3.30 GFLOPS
                    2011-03-24 21:11:38.5347 [PID=9363 ] [send] est. duration for WU 95079037: unscaled 42400.32 scaled 67631.24
                    2011-03-24 21:11:38.5347 [PID=9363 ] [HOST#2335620] Sending [RESULT#224692340 PM0068_03441.dm_388_1] (est. dur. 67631.24 seconds)
                    2011-03-24 21:11:38.5360 [PID=9363 ] [send] don't need more work
                    2011-03-24 21:11:38.5362 [PID=9363 ] [send] don't need more work
                    2011-03-24 21:11:38.5362 [PID=9363 ] [send] don't need more work
                    2011-03-24 21:11:38.5362 [PID=9363 ] [send] don't need more work
                    2011-03-24 21:11:38.5372 [PID=9363 ] Sending reply to [HOST#2335620]: 5 results, delay req 60.00
                    2011-03-24 21:11:38.5376 [PID=9363 ] Scheduler ran 0.040 seconds

                    alephnull
                    alephnull
                    Joined: 2 Dec 05
                    Posts: 21
                    Credit: 28551947
                    RAC: 0

                    richard, i did receive

                    24 Mar 2011 22:29:08 UTC
                    Message 104770 in response to message 104769
                    (moderation:
                      )

                      richard,

                      i did receive those wu. after waiting a bit, i got these subsequent messages:

                      859 Einstein@Home 2011-03-24 17:22:59 Starting PM0068_03451.dm_12_0
                      860 Einstein@Home 2011-03-24 17:23:00 Starting task PM0068_03451.dm_12_0 using einsteinbinary_BRP3 version 105
                      861 Einstein@Home 2011-03-24 17:25:07 Starting PM0068_03441.dm_420_1
                      862 Einstein@Home 2011-03-24 17:25:08 Starting task PM0068_03441.dm_420_1 using einsteinbinary_BRP3 version 105
                      863 Einstein@Home 2011-03-24 17:42:42 Starting PM0068_03441.dm_388_1
                      864 Einstein@Home 2011-03-24 17:42:42 Starting task PM0068_03441.dm_388_1 using einsteinbinary_BRP3 version 105
                      865 Einstein@Home 2011-03-24 17:50:08 Starting PM0068_03441.dm_384_0
                      866 Einstein@Home 2011-03-24 17:50:08 Starting task PM0068_03441.dm_384_0 using einsteinbinary_BRP3 version 105
                      867 Einstein@Home 2011-03-24 18:13:21 Starting PM0057_027B1.dm_44_2
                      868 Einstein@Home 2011-03-24 18:13:21 Starting task PM0057_027B1.dm_44_2 using einsteinbinary_BRP3 version 105
                      ...
                      872 Einstein@Home 2011-03-24 18:14:56 Sending scheduler request: To fetch work.
                      873 Einstein@Home 2011-03-24 18:14:56 Requesting new tasks for GPU
                      874 Einstein@Home 2011-03-24 18:14:59 Scheduler request completed: got 1 new tasks
                      875 Einstein@Home 2011-03-24 18:15:01 Started download of PM0068_034D1_420.binary
                      876 Einstein@Home 2011-03-24 18:15:01 Started download of PM0068_034D1_421.binary
                      877 Einstein@Home 2011-03-24 18:15:16 Finished download of PM0068_034D1_420.binary
                      878 Einstein@Home 2011-03-24 18:15:16 Finished download of PM0068_034D1_421.binary
                      879 Einstein@Home 2011-03-24 18:15:16 Started download of PM0068_034D1_422.binary
                      880 Einstein@Home 2011-03-24 18:15:16 Started download of PM0068_034D1_423.binary
                      881 Einstein@Home 2011-03-24 18:15:29 Finished download of PM0068_034D1_422.binary
                      882 Einstein@Home 2011-03-24 18:15:30 Finished download of PM0068_034D1_423.binary

                      removed non einstein messages from log. now theres one einstein gpu wu now. let me know if i can provide any other information.

                      rob

                      Comment viewing options

                      Select your preferred way to display the comments and click "Save settings" to activate your changes.

                      Forums › Cruncher's Corner