Hi Jinal:
Jinal Jhaveri wrote:
> Hi All,
>
> I am using load_avg=2.0 as a load threshold for our SGE
nodes. I avoid
> using the np_load_avg (load per processor) as it is
mis-interpreted on
> intel machines with hyperthreading enabled. Can anyone
(especialy in
> Bioinformatics setting running jobs like
blast,interpro) share their
> experience on what load threshold works best for them?
Unless you have a really good reason for running HT, I would
probably
suggest disabling it. Our tests with BLAST a few years ago
showed no
real benefit. You do get benefit from bigger caches and
faster memory
systems, but not HT.
With that in mind, I would probably stick to the load
average == number
of cores in system.
Joe
>
> Thank you!
>
> --Jinal
>
> ps. Apologies for cross posting but I realized that
both SGE gurus and
> Bio-cluster people might be able to answer this
question.
>
>
>
>
>
>
>
------------------------------------------------------------
---------
> To unsubscribe, e-mail: users-unsubscribe gridengine.sunsource.net
> For additional commands, e-mail: users-helpgridengine.sunsource.net
--
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landmanscalableinformatics.com
web : http://www.scalabl
einformatics.com
phone: +1 734 786 8423
fax : +1 734 786 8452 or +1 866 888 3112
cell : +1 734 612 4615
_______________________________________________
Bioclusters maillist - Bioclustersbioinformatics.org
https://bioinformatics.org/mailman/listinfo/bioclusters
|