WU's too small for GPU
log in

Advanced search

Message boards : Windows : WU's too small for GPU

Previous · 1 · 2 · 3
Author Message
Send message
Joined: 28 Nov 16
Posts: 19
Credit: 5,313,490
RAC: 0
Message 289 - Posted: 27 Jun 2017, 18:51:49 UTC - in response to Message 271.

Now running some 1.04 WU's, they seem to be running even smoother than the 1.02/03 WU's....

Sorry, I hurried to much to release them and skipped some routine testing. They produce invalid results on some machines and I rolled back to version 1.03 for now.

I see we are up at version 1.05 now. Back to running like it was before smooth and stable at 5x. In fact in testing I had it up to 7x before it started showing instability. Of course at 7x it runs a lot slower, but it was stable.

One thing I've noticed, in the app_config file, the line to regulate the CPU percentage, doesn't have any effect on the app in Boinc manager at all. Zero none. I have set the CPU ranging from .2 to 1.0 and it has no effect at all.

I'm using Process Lasso to set the CPU usage and it appears to take two cores/threads for every GPU. In my case that would be ten cores/threads. (two cores x five parallel WU per GPU)

The funny part is Process Lasso shows it only using five of the allotted cores/threads. And, since the processes are using only one core per five GPU processes, so since it is only using five cores in actuality, when I cut Process Lasso to five cores it throttles the process down to about half speed.

It needs the two cores per GPU. This is on the Intel DP, Also it doesn't matter if it is a physical core or a logical core. It will use either, but it HAS to have 2 per GPU.

On the other hand, the FX8350 boxes don't have this behavior. They operate like normal and the app_config file works as normal, the CPU line works to limit CPU usage per core.

Interesting to say the least.

I wonder what is causing the difference?

For GPU projects the cpu usage only tells BOINC Mgr how many CPU cores to reserve for the GPU tasks. It does not change how much CPU usage the gpu exe's actually consume.

An example: This will reserve 1 CPU core to do all the GPU work. If you have 8 CPU threads available then 7 CPU BOINC tasks can run. The GPUs may take 1 CPU thread each so in actuality there may only be 5 CPU threads to do 7 CPU tasks.

<app_config> <app> <name>xansons_gpu</name> <gpu_versions> <gpu_usage>0.33</gpu_usage> <cpu_usage>0.33</cpu_usage> </gpu_versions> </app> </app_config>

Project administrator
Project developer
Project tester
Project scientist
Help desk expert
Send message
Joined: 26 Oct 16
Posts: 321
Credit: 103,382
RAC: 0
Message 292 - Posted: 27 Jun 2017, 19:10:23 UTC - in response to Message 271.
Last modified: 27 Jun 2017, 19:17:45 UTC

It needs the two cores per GPU. This is on the Intel DP, Also it doesn't matter if it is a physical core or a logical core. It will use either, but it HAS to have 2 per GPU.

If you want to significantly reduce the CPU load in GPU tasks, you can specify:

in your app_config.xml. But this will completely freeze your screen considering that you run multiple GPU tasks in parallel. So make sure that the option to suspend GPU computing when computer is in use is selected, otherwise it will be hard even to stop the computing.

Update. This should be specified under app_version. So if you are running 4 tasks per GPU, your app_config should contain:
<app_version> <app_name>xansons_gpu</app_name> <plan_class>opencl_ati_102_windows</plan_class> <ngpus>0.25</ngpus> <cmdline>--nowait</cmdline> </app_version>

It is not necessary to specify this option for Nvidia.

Send message
Joined: 22 Jun 17
Posts: 15
Credit: 12,727,586
RAC: 0
Message 293 - Posted: 28 Jun 2017, 2:23:37 UTC - in response to Message 292.

I think you missed what I was saying, but no matter.....

I think I understand how to set up a boinc client, I've done it a couple of times.

I applaud your willingness to work with your crunchers, that is a rarity in the boinc project world.

Beautiful project rollout!

Good luck!

Previous · 1 · 2 · 3

Message boards : Windows : WU's too small for GPU

Main page · Your account · Message boards

© 2020 Vladislav Neverov (NRC 'Kurchatov institute'), Nikolay Khrapov (Institute for Information Transmission Problems of RAS)