I think they were also asking for GPU throttling and as far as I know, it can't be done other than to change the number of local and global threads such that a low end GPU is responsive. Unfortunately, that means the high end GPUs are only running at 80% or less. That's where anonymous apps come in so users can change the settings so that their machine works as fast as possible but not to the point where the system is unresponsive.
On Sat, Jan 12, 2013 at 5:04 PM, David Anderson <[email protected]>wrote: > Previously, if a job used less than 1 CPU, > it was not subject to CPU throttling. > > With my change: GPU jobs are subject to CPU throttling > no matter how many CPUs they use. > > (Note: GPU apps rely on the CPU to "feed" the GPU, > usually many times per second; so when we suspend the CPU > part of the app, we effectively suspend the GPU part too). > > ---- > > The request for a separate GPU throttling preference is noted; > however we're not going to get to it immediately. > > -- David > > On 12-Jan-2013 2:17 PM, Jacob Klein wrote: > > Can you explain your change a little more, to further explain what it > does? > > > > I checked in a change so that CPU throttling applies to GPU apps also. > > > Can anyone think of a reason not to do this? > > > -- David > _______________________________________________ > boinc_alpha mailing list > [email protected] > http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_alpha > To unsubscribe, visit the above URL and > (near bottom of page) enter your email address. > _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
