Hi,

I've noticed that when Milky-way uses one of my NVIDIA boards, it uses all of it (1 GPU). In Seti@Home, I'm able to constrain or divide GPU workload among multiple work units. How can I do the same with Milky-way?

Thanks,
Jim Wright

The same way as at SETI - use an application configuration file.

So far as I know, Einstein@Home is the only project which enables you to control GPU usage via your account settings on their website.

Richard,

Thanks for the hint. Now, where do I get a sample application config file customized for MilkyWay?

Jim Wright

Hi,

Does anyone have a properly coded Application Config file for Milky Way? Please share it?

I do not feel competent to construct one from scratch and really really want to change the CUDA processing parameters to improve my overall system performance.

Thanks,
Jim Wright

Hi,

Does anyone have a properly coded Application Config file for Milky Way? Please share it?

I do not feel competent to construct one from scratch and really really want to change the CUDA processing parameters to improve my overall system performance.

Thanks,
Jim Wright

Post your Seti one, maybe someone can help you modify it?

I found this on the updated GPU requirements thread; you may have to change the names of the executables to match those in your Program Data/BOINC/Projects/Milkyway... directory:

http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3188&postid=62765

<app_config>

<app>
<name>milkyway</name>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.5</cpu_usage>
</gpu_versions>
</app>

<app>
<name>milkyway_separation__modified_fit</name>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.5</cpu_usage>
</gpu_versions>
</app>

</app_config>