On Mon, 18 Dec 2006, nate lipkowitz wrote:
which is pretty bad linearity with respect to the number of processors involved. however this is also a pretty small job. is this normal for mpb-mpi calculations involving small jobs? can i expect an improvement with larger jobs, or should i dig into locally-specific issues like interconnects or scheduling priorities and whatnot?
Yes, you should get more of an improvement with larger jobs. Generally, I only use mpb-mpi for very large calculations, especially defect mode calculations.
If you want to parallelize small calculations you should look into mpb-split, which runs the different k points in parallel (see the manual).
Steven _______________________________________________ mpb-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/mpb-discuss
