Bioperl: off topic Beowulf question
Nathan O. Siemers
Fri, 03 Mar 2000 18:31:07 -0500
We've had a cluster running blast for almost 6 months now. We don't
use it as a backend for a web blast page, but for batch computation.
We have had good experiences with the free PBS queuing system. No
special problems compiling the ncbi toolkit for the cluster, other
than the usual headaches (remove parts of the makefile that need
It rocks. The major limitation is linux kernel file size
limitations, as the databases are getting huge of course.
Keith Allen <firstname.lastname@example.org> writes:
> Hey folks,
> sorry to wade in with an off topic question, but I know this stuff
> central to what a lot of you think about, at least from time to time.
> considering putting together a Beowulf cluster as a dedicated Blast
> but we'd really like to talk to someone who's already done this to find
> what's worked well and what hasn't. Specifically we need to know how
> people are handling job scheduling and load balancing, what sort of
> measures were needed to compile Blast for the cluster, and any other
> tidbits anyone might have to share.
> We're planning to put NCBI Blast on this server, but we're very
> to know if there are folks who think WU-Blast is better suited.
> -Keith Allen
> Bioinformatics Scientist
> Paradigm Genetics
> RTP, NC
> =========== Bioperl Project Mailing List Message Footer =======
> Project URL: http://bio.perl.org/
> For info about how to (un)subscribe, where messages are archived, etc:
N a t h a n O . S i e m e r s
Division of Applied Genomics
Bristol-Myers Squibb Pharmaceutical Research Institute
Hopewell Building 3B, P.O. Box 5400, Princeton, NJ 08543-5400
=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org/
For info about how to (un)subscribe, where messages are archived, etc: