On Wed, 2 Mar 2005, kalyani goli wrote: > Iam running this on SGI multiprocessor(numa). So, everything is > visible to all the processors. Thats why i gave the directory where > ncbi is installed ,and gave it same for both shared and local. ... gurgh? mpiblast will (and I have seen it, with my own eyes) scale 'super linear' on high latency, multiply interconnected servers. Aaron did a really sensible and good job to make this scale across nodes. However, as you describe: "Iam running this on SGI multiprocessor(numa)", you are running on a single shared (well near unified, and SGI do this very, very well) memory server with, as you said and appear to understand shared storage... *sigh* What on earth are you going to gain from MPI? Standard NCBI threads should do for you just fine, or maybe I've been smoking the funny stuff. If you _do_ happen to have multiple NUMA's in a cluster, (1) you are very lucky and (2) you should the still listen to Joe's advice... Local is only local so far, try: Shared=/home/kalyani/toolkit/ncbi Local=/tmp/kalyani_mpiblast/ (or as Joe maybe put better) Shared=/home/kalyani/toolkit/ncbi Local=/mylocalfilesystemthatnoonewillmesswith/kalyani_mpiblast/ WFM, YMMV.. Best, J. -- James Cuff, D. Phil. Group Leader, Applied Production Systems Broad Institute of MIT and Harvard. 320 Charles Street, Cambridge, MA. 02141. Tel: 617-252-1925 Fax: 617-258-0903