Hi Guys, I tried to run clump finder using the sample script available on web and just changed the minimum value of the density. Running on single cpu took too much time. So, it tried to run it in parallel. Yt did not divide the task among cpus properly and just executed the same process on all cpus. In this way, it consumed all the memory available on the machine (128 GB ). Am i doing some thing stupid or there is a problem with yt clump finder? mpirun -np 48 ./clumpfind.py --parallel Thanks in advance! Latif
Hi Latif, You weren't doing anything stupid. Unfortunately, the clump finder does not work in parallel. yt is currently lacking a contour finding algorithm that works in parallel. If anyone is interested in writing one, they would be very welcome! Britton On Mon, Aug 13, 2012 at 1:23 PM, Latif <latifne@gmail.com> wrote:
Hi Guys, I tried to run clump finder using the sample script available on web and just changed the minimum value of the density. Running on single cpu took too much time. So, it tried to run it in parallel. Yt did not divide the task among cpus properly and just executed the same process on all cpus. In this way, it consumed all the memory available on the machine (128 GB ). Am i doing some thing stupid or there is a problem with yt clump finder? mpirun -np 48 ./clumpfind.py --parallel Thanks in advance! Latif
_______________________________________________ yt-users mailing list yt-users@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
participants (2)
-
Britton Smith
-
Latif