New Improved Gravational Wave App - Discussion

Forums › Cruncher's Corner

Tom M
Tom M
Joined: 2 Feb 06
Posts: 6432
Credit: 9560985908
RAC: 10218205

Yavanius wrote: Ian&Steve C.

29 Jan 2024 23:53:08 UTC
Message 221824 in response to message 221812
(moderation:
    )

    Yavanius wrote:

    Ian&Steve C. wrote:

    boinc identifies your GPU as only having 2GB of VRAM, which is not enough to get the GW tasks, you need at least 3GB

     

    Ah...I see. I wonder why I was thinking it had 4gb... Must have been thinking of something else...wishful thinking? ^_^

    https://www.techpowerup.com/gpu-specs/geforce-gtx-960m.c2635

    Thinks it has 4 GB of memory.  Maybe there is a setting in the Bios for "how big" the gpu memory is?

    Tom M

    A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)  I want some more patience. RIGHT NOW!

    Gandolph1
    Gandolph1
    Joined: 20 Feb 05
    Posts: 180
    Credit: 389642764
    RAC: 870

    Tom M wrote: Yavanius

    30 Jan 2024 1:02:30 UTC
    Message 221826 in response to message 221824
    (moderation:
      )

      Tom M wrote:

      Yavanius wrote:

      Ian&Steve C. wrote:

      boinc identifies your GPU as only having 2GB of VRAM, which is not enough to get the GW tasks, you need at least 3GB

       

      Ah...I see. I wonder why I was thinking it had 4gb... Must have been thinking of something else...wishful thinking? ^_^

      https://www.techpowerup.com/gpu-specs/geforce-gtx-960m.c2635

      Thinks it has 4 GB of memory.  Maybe there is a setting in the Bios for "how big" the gpu memory is?

      Tom M

      You could also use GPU-Z from https://www.techpowerup.com/download/techpowerup-gpu-z/ and see what it says the total memory of your card is....

       

      Ian&Steve C.
      Ian&Steve C.
      Joined: 19 Jan 20
      Posts: 3945
      Credit: 46598002642
      RAC: 64172479

      scroll to the list at the

      30 Jan 2024 1:04:33 UTC
      Message 221828 in response to message 221824
      (moderation:
        )

        scroll to the list at the bottom of that page and you'll see there are some 2GB variants as well. his is probably 2GB. laptop Nvidia discrete GPUs don't usually (or ever?) have shared memory with the CPU to which it can be adjusted in the BIOS. They have their own memory.

        _________________________________________________________________________

        Tom M
        Tom M
        Joined: 2 Feb 06
        Posts: 6432
        Credit: 9560985908
        RAC: 10218205

        Ian&Steve C. wrote: scroll

        30 Jan 2024 3:30:55 UTC
        Message 221830 in response to message 221828
        (moderation:
          )

          Ian&Steve C. wrote:

          scroll to the list at the bottom of that page and you'll see there are some 2GB variants as well. his is probably 2GB. laptop Nvidia discrete GPUs don't usually (or ever?) have shared memory with the CPU to which it can be adjusted in the BIOS. They have their own memory.

          Ah, sorry.

          A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)  I want some more patience. RIGHT NOW!

          Tom M
          Tom M
          Joined: 2 Feb 06
          Posts: 6432
          Credit: 9560985908
          RAC: 10218205

          I am trying to run

          31 Jan 2024 2:31:17 UTC
          Message 221868
          (moderation:
            )

            ===edit===

            Never mind.  Just noticed a bunch of All-Sky computation errors for today.  So have switched brp7 to single task.  And have started downloading All-Sky only again.

            ===edit===

             

            I am trying to run brp7/meerKat at the same time as All-Sky GW on one gpu.

            Here is what my app_config.xml looks like.  It is currently running a single MeerKat task.  And in theory should be able to run 2 All-Skys.

            What am I missing.  Or is it just not possible?

            ==========

            <app_config>
            <app>
            <name>einstein_O3AS</name>
            <max_concurrent>5</max_concurrent>
            <gpu_versions>
            <gpu_usage>0.333</gpu_usage>
            <cpu_usage>0.9</cpu_usage>
            </gpu_versions>
            </app>
            <app>
            <name>einsteinbinary_BRP7</name>
            <max_concurrent>1</max_concurrent>
            <gpu_versions>
            <gpu_usage>0.333</gpu_usage>
            <cpu_usage>0.9</cpu_usage>
            </gpu_versions>
            </app>
            <app>
            <name>hsgamma_FGRP5</name>
            <max_concurrent>1</max_concurrent>
            </app>

            </app_config>

            =============

            NM

            A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)  I want some more patience. RIGHT NOW!

            Tom M
            Tom M
            Joined: 2 Feb 06
            Posts: 6432
            Credit: 9560985908
            RAC: 10218205

            Go my first NVIDIA Titan V in

            1 Feb 2024 3:20:24 UTC
            Message 221919
            (moderation:
              )

              Go my first NVIDIA Titan V in Monday.  Its finally online now.

              Yes, it is a LOT quieter than my rtx 3080 ti Turbo blower gpus.

              Since it is currently in "test only" mode I have been running a some tasks then bumping the number of tasks running at the same time on it.

              1x is ~24 minutes

              2x is ~27 minutes

              These were VERY small samples.

              3x has varied from 28-35 minutes. 

              I will be running it until bedtime tonight since it is not a production system (yet.).

              Tom M

               

              A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)  I want some more patience. RIGHT NOW!

              Boca Raton Community HS
              Boca Raton Comm...
              Joined: 4 Nov 15
              Posts: 238
              Credit: 10505475586
              RAC: 27049469

              Tom M wrote:Go my first

              1 Feb 2024 5:03:14 UTC
              Message 221921 in response to message 221919
              (moderation:
                )

                Tom M wrote:

                Go my first NVIDIA Titan V in Monday.  Its finally online now.

                Yes, it is a LOT quieter than my rtx 3080 ti Turbo blower gpus.

                Since it is currently in "test only" mode I have been running a some tasks then bumping the number of tasks running at the same time on it.

                1x is ~24 minutes

                2x is ~27 minutes

                These were VERY small samples.

                3x has varied from 28-35 minutes. 

                I will be running it until bedtime tonight since it is not a production system (yet.).

                Tom M

                 

                That's awesome that you were able to get one! 

                Also, as promised, I am wrapping up the "study" of how many concurrent work units are ideal for this work (for us). I should have all the results posted tomorrow (hopefully). The last group is running right now (10x) and it has been fun to run these tests. Stay tuned!

                Tom M
                Tom M
                Joined: 2 Feb 06
                Posts: 6432
                Credit: 9560985908
                RAC: 10218205

                Boca Raton Community HS

                1 Feb 2024 11:55:37 UTC
                Message 221926 in response to message 221921
                (moderation:
                  )

                  Boca Raton Community HS wrote:

                  That's awesome that you were able to get one! 

                  Also, as promised, I am wrapping up the "study" of how many concurrent work units are ideal for this work (for us). I should have all the results posted tomorrow (hopefully). The last group is running right now (10x) and it has been fun to run these tests. Stay tuned!

                  I have my guitar around here someplace.  Its been a while.  So I will have to tune it first.  Then I can "Stay tuned!"  

                  :)

                  I do look forward to more data.

                  Tom M

                  A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)  I want some more patience. RIGHT NOW!

                  Boca Raton Community HS
                  Boca Raton Comm...
                  Joined: 4 Nov 15
                  Posts: 238
                  Credit: 10505475586
                  RAC: 27049469

                  I just finished the analysis

                  1 Feb 2024 17:41:55 UTC
                  Message 221936
                  (moderation:
                    )

                    I just finished the analysis of concurrent running of these tasks. I cannot say there is anything landmark within the results but maybe some interesting insights. Also, when (inevitably) users ask what the ideal amount of concurrent tasks are for these work units, this link can just be referred to. This is intended to be an informal look at the concurrent running of these tasks and not an actual study that would require tighter controls. 

                    Of course, argument over the validity of the statistical analyses chosen for this is welcomed but I cannot say I am in any way an expert in statistics. 

                    I am just going to post the link to the report since there are figures, graphs, etc. 

                    This was a lot of fun to do and I hope it helps someone out there. 

                    https://docs.google.com/document/d/1X9WVmkTBfqbMzw5nKLys2xeYU8eruAUXrkYiUJx-iN4/edit?usp=sharing

                    Tom M
                    Tom M
                    Joined: 2 Feb 06
                    Posts: 6432
                    Credit: 9560985908
                    RAC: 10218205

                    Thank you.+42

                    1 Feb 2024 22:20:54 UTC
                    Message 221941 in response to message 221936
                    (moderation:
                      )

                      Thank you.

                      Nice.

                      +42

                       

                      ===edit==

                      So for my purposes 3x,4x or 5x are possibly my best choices.

                      I believe I am memory constrained at 5x.

                      Only testing on the rtx 3080 ti and the Titan V will tell me which.

                      A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)  I want some more patience. RIGHT NOW!

                      Comment viewing options

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

                      Forums › Cruncher's Corner