[BiO BB] Anybody knows the usage stats for different apps in sequence alignment & molecular dynamics?
Dan Bolser
dmb at mrc-dunn.cam.ac.uk
Thu Jan 15 10:05:38 EST 2004
++ CCGB--
>
>> Our misison is to enable individual researchers and small
>> labs to speed up their research and level the playing field with the 'big
>> guys' who can afford throwing 100K per one
>> acceleration module...
>
> My impression is that computational speed is the limiting factor for only a small
> number of groups. I think that most researchers remain interface and data
> management bound.
>
> I could be way off base with this, and I'm curious to hear about other
> experiences: I believe that (as is usually true with high performance computing)
> 5% of the researchers take up 95% of the cluster time, and the other 95% are doing
> the best they can to manage even a rather small set of results.
>
> Am I wrong?
No you are not wrong, but perhaps we are just privilaged? Anyone want to borrow my
idle time?
Most of my time is scripting / debugging jobs to run. The machines sit idel most of
the time, and finish in reasonable time for everything I need.
I would be more likely to pay for a process pipeline manager - naturally I would
preffer a free one.
Any projects to make a process pipeline tool in php?
Should -
* allow workflows to be created / edited / monitored through web.
* be tied into cron.
* rely on gnu make under the hood.
* be tied into cvs for 'job module' developement.
* provide error reporting by mail.
* provide log review.
> Are lots of people sitting on their hands waiting for BLAST (and
> HMMER, and InterPro, and meta-java-dock-o-matic-whatever) to
> complete?
>
> -Chris Dwan
> The University of Minnesota
> _______________________________________________
> BiO_Bulletin_Board maillist - BiO_Bulletin_Board at bioinformatics.org
> https://bioinformatics.org/mailman/listinfo/bio_bulletin_board
More information about the BBB
mailing list