Skip to first unread message
McCoy
Jan 6, 2017, 10:48:49 AM1/6/17
to Iris
Hi there,
I am reading in largish (~5Gb) .pp files and regridding them to a more manageable size. This seems to take quite a while, even though the process should be embarrassingly parallel. I am working on the high memory node on Jasmin and the CPU stays around 70-90%. Is there a way to improve performance?
The code in question is:
Z=Z.regrid(regridder,iris.analysis.AreaWeighted(mdtol=1))
where Z is a cube in space and time. Regridder is a cube with the coarse grid.
Thanks,
I am reading in largish (~5Gb) .pp files and regridding them to a more manageable size. This seems to take quite a while, even though the process should be embarrassingly parallel. I am working on the high memory node on Jasmin and the CPU stays around 70-90%. Is there a way to improve performance?
The code in question is:
Z=Z.regrid(regridder,iris.analysis.AreaWeighted(mdtol=1))
where Z is a cube in space and time. Regridder is a cube with the coarse grid.
Thanks,
Reply all
Reply to author
Forward
0 new messages