[Bioclusters] Local blast server, beowulf vs mosix + BIOITWORLD conference

chris dagdigian bioclusters@bioinformatics.org
Tue, 05 Mar 2002 11:16:51 -0500


Great to see lots of new subscribers to the list _and_ a great thread. I 
think this list is starting to reach critical mass.

Three things:

(1) My limited experience with MOSIX has me believing that in the MOSIX 
world a process that is doing heavy I/O operations will never get 
migrated across to a less loaded machine. This alone is enough for me to 
not consider MOSIX/SSI for bioclusters because all of the ones I have 
built so far are very, very often used for IO-bound embarassingly 
parallel jobs (blast, genscan, etc. etc). Am I totally wrong? Anyone out 
there using SSI/Mosix systems for hardcore biology stuff?

(2) Diskless. I would never _not_ put a large cheap IDE drive in my 
cluster elements because at the very least I need a local large /tmp or 
/scratch partition for caching sequence databases and raw data. Without 
this approach you pretty much will swamp your NFS server under any sort 
of serious workload. Diskless I guess would be cool for limited-scope 
systems that will only be doing crazy MPI/PVM or CPU-bound jobs. Anyone 
doing this?

Also much of the benefits touted by proponants of diskless clusters such 
as ease of cluster management and ease of OS image management go away if 
you use something like SystemImager or the FAI scripts that were 
referred to in a previous  email. Are there any other advantages to 
diskless clusters that people have seen for real world life science 
computing? Ram disks?



(3) BIOITWORLD Roll Call

Anyone coming to Boston for the IDG BioITWorld conference? I was 
supposed to organize the biocluster BOF along with Andrew Fant, Glen 
Otero and Jeff Bizzaro but it looks as though I'm going to be on the 
west coast that week. Glen, Jeff & Andy will run the BOF without me and 
we'll post the details to this list once we have them.

Give a shout if you are coming to town. It would be nice to know how 
many list members will be in attendence.

Also- any other conferences/workshops/meetings coming up that people 
recommend?


-Chris



-- 
Chris Dagdigian, <dag@sonsorol.org>
Life Science IT & Research Computing Freelancer &
member of the bioteam.net consulting collective
Office: 617-666-6454, Mobile: 617-877-5498, Fax: 425-699-0193
Yahoo IM: craffi