[Bioclusters] Re: BLAST job on SGE

[Jose Duarte]

> > We are running SGE on an Apple cluster. When an array of BLAST jobs are 
> > submitted, it schedules every two jobs to one compute node, regardless of how 
> > many compute nodes are available or how many jobs. Any one knows how to adjust 
> > it to schedule one job to one node each time? 

There is a suggested approach to running one job per node on this thread
from the SGE mailing list (works only on SGE 6):

http://gridengine.sunsource.net/servlets/ReadMsg?msgId=12107&listName=users

We've tried this successfully here. Jobs get scheduled at one per node
if sent to the special 1-slot-per-node queue. Other 2-slots-per-node
queues are subordinated to this one so jobs sent to these queues get
scheduled in the default way, but waiting first for anything running on
the 1-slot-per-node queue.

In the case of blast, it can also be used in combination with the "-a"
option to use 2 processors for the blast run.

Cheers

Jose


---
Jose M. Duarte
MRC Functional Genetics Unit
Dept of Human Anatomy & Genetics
University of Oxford
South Parks Road
Oxford OX1 3QX
UK