Message boards :
Number crunching :
nbody: Trying to run on 2 cores, but downloading 4 core tasks
Message board moderation
Author | Message |
---|---|
Send message Joined: 8 Jan 18 Posts: 44 Credit: 43,798,833 RAC: 2,383 |
So I've decided to start running the Nbody tasks again. I have a 4 core CPU that has 2 GPUs. I want to limit Nbody to run on 2 cores maximum so my GPUs stay busy. I added the following to my app_config file (my whole app_config included, just in case, but the nbody section is what I added): <app_config> <app> <name>milkyway</name> <gpu_versions> <gpu_usage>0.5</gpu_usage> <cpu_usage>0.5</cpu_usage> </gpu_versions> </app> <app_version> <app_name>milkyway_nbody</app_name> <plan_class>mt</plan_class> <cmdline>--nthreads 2</cmdline> </app_version> </app_config> However, when I reloaded my app_config, I have downloaded 4 CPU Nbody tasks. I'm missing something obvious, I just don't know what it is. Any ideas? On a side note, I think I observed these 4 CPU Nbody tasks running while my 2 GPUs were also running...so 4 GPU tasks (using a sum of 2 CPUs), and a 4 CPU task. The Nbody task seemed to run slow, but once I suspended GPU tasks, the nbody sped up. I haven't had a chance to experiment with it more, but I'm curious if anyone else has noticed this. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
So I've decided to start running the Nbody tasks again. I have a 4 core CPU that has 2 GPUs. I want to limit Nbody to run on 2 cores maximum so my GPUs stay busy. I added the following to my app_config file (my whole app_config included, just in case, but the nbody section is what I added): N-body tasks are cpu only and by default are multi-threading while the gpu tasks are Seperation tasks and by default use some part of a gpu depending on the gpu. 0.5 times 4 equals 2 cores, that leaves zero cores for your gpu's to use so of course the cpu units will speed up if the gpu's stop. Also specifing 0.5 cpu usage is only a suggestion, the app that the Project writes is the determining factor of how much of a cpu core it will use. Most do not use a full core so it's not usually a problem but in some cases it is and no app or whatever file can overwrite it. |
Send message Joined: 8 Jan 18 Posts: 44 Credit: 43,798,833 RAC: 2,383 |
Mikey, the section of app_config I just added was everything contained in the <app_version> section. Everything else was there prior to me re-engaging with nbody. I don't understand why I have specified 2 cores max for nbody, but for some reason I'm still downloading 4 core tasks. I don't think this has nothing to do with my GPUs nor the separation tasks. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
Mikey, the section of app_config I just added was everything contained in the <app_version> section. Everything else was there prior to me re-engaging with nbody. I don't understand why I have specified 2 cores max for nbody, but for some reason I'm still downloading 4 core tasks. I don't think this has nothing to do with my GPUs nor the separation tasks. Probably because Boinc is pre-fetching tasks so you don't run out of work, the newer versions of Boinc do that even with a resource share of zero. |
Send message Joined: 24 Jan 11 Posts: 715 Credit: 556,092,420 RAC: 51,540 |
So I've decided to start running the Nbody tasks again. I have a 4 core CPU that has 2 GPUs. I want to limit Nbody to run on 2 cores maximum so my GPUs stay busy. I added the following to my app_config file (my whole app_config included, just in case, but the nbody section is what I added): Try this version of your app_config. <app_config> <app> <name>milkyway</name> <gpu_versions> <gpu_usage>0.5</gpu_usage> <cpu_usage>0.5</cpu_usage> </gpu_versions> </app> <app_version> <app_name>milkyway_nbody</app_name> <max_concurrent>1</max_concurrent> <plan_class>mt</plan_class> <avg_ncpus>2</avg_ncpus> <cmdline>--nthreads 2</cmdline> </app_version> </app_config> |
Send message Joined: 8 Jan 18 Posts: 44 Credit: 43,798,833 RAC: 2,383 |
Keith, no dice. After inputting your suggestion, I set MW@H to NNT, ran through all CPU tasks (I had no other tasks from other projects, fwiw). Once I started accepting new tasks again I kept getting 4 core nbody. |
Send message Joined: 8 May 09 Posts: 3339 Credit: 524,010,781 RAC: 0 |
Keith, no dice. After inputting your suggestion, I set MW@H to NNT, ran through all CPU tasks (I had no other tasks from other projects, fwiw). Once I started accepting new tasks again I kept getting 4 core nbody. Try this....set MW to no new tasks and get rid of all the n-body tasks by either aborting them or running them, then set your pc to use only 50% of the cpu's and allow more n-body tasks and see if they don't come in as 2 core tasks. Once your cache is full turn off all new n-body tasks again and change the pc to use all 4 cpu cores and your pc should happily crunch the n-body tasks with only 2 cpu cores. Yes it's a pain in the butt but it should work until you can figure out the right app_config file setting. |
Send message Joined: 8 Jan 18 Posts: 44 Credit: 43,798,833 RAC: 2,383 |
I could try that, but I am running other projects, and I'm using all four cores. I haven't limited my core usage before on this computer and I don't want to start now. Additionally, I don't have the time to babysit this computer.Keith, no dice. After inputting your suggestion, I set MW@H to NNT, ran through all CPU tasks (I had no other tasks from other projects, fwiw). Once I started accepting new tasks again I kept getting 4 core nbody. Is there a list of command line options for this application? I know SETI@Home did with their MB and AP applications. |
Send message Joined: 9 Jul 17 Posts: 100 Credit: 16,967,906 RAC: 0 |
So I've decided to start running the Nbody tasks again. I have a 4 core CPU that has 2 GPUs. I want to limit Nbody to run on 2 cores maximum so my GPUs stay busy. This works for me for the CPU. You can add the GPU part. You will need a reboot (or restarting BOINC) for it to take effect for any work already downloaded. <app_config> <app> <app_version> <app_name>milkyway_nbody</app_name> <plan_class>mt</plan_class> <avg_ncpus>2</avg_ncpus> </app_version> </app_config> |
Send message Joined: 8 Jan 18 Posts: 44 Credit: 43,798,833 RAC: 2,383 |
Unfortunately, this did not work. Steps to reproduce: 1. BOINC running. MW set to NNT. 2. Edit app_config as Jim1348 mentions above. Save and close. 3. Continue running BOINC until all MW tasks are completed. I did not reload the app_config file at this time. 4. Once all MW tasks have been returned, exited BOINC. 5. After waiting a minute, re-loaded BOINC. 6. Set MW to receive tasks, and 4-core Nbody tasks were downloaded. Additionally, I still have the 0.5 CPU/0.5 GPU setting for the separation tasks. When I started recieving new files again, it is running five tasks total (4 separation tasks, 2 per GPU), and one nbody task. That doesn't seem right, either. |
Send message Joined: 16 Mar 10 Posts: 213 Credit: 108,397,914 RAC: 3,422 |
Unfortunately, this did not work. Steps to reproduce: It seems right to me: what you may not realise is that the server doesn't know you've tweaked the client to only use 2 cores for an Nbody job so it sends it out tagged as using 4 cores. But when your client actually runs an Nbody task with your revised configuration, there are now two cores left over and you're getting 4 Separation GPU jobs in those slots. Unless it suddenly stops leaving 2 cores free for other work, I would assume it's doing what you want... Cheers - Al. |
©2024 Astroinformatics Group