[Bioclusters] Overcoming Grid drawbacks

Kathleen kathleen at massivelyparallel.com
Sat Jan 21 00:22:23 EST 2006

Thanks Chris.  Ok...a reality check.  I'm relatively new to this industry
and I have a huge...I mean really huge learning curve.   I not only need to
understand bioinformatics, but I also need to understand supercomputing and
my company's advantage.  I'm a newbie (just started with MPT last Sept.) I
really do want to follow proper etiquette.  I really am interesting in
finding out what matters to users.  I'll just leave it at that.  No
marketing...I just want to know what users need.  I think we can help, but
I'll lay that aside to find out more about what users need.


Profile: https://www.linkedin.com/e/fps/3301509/
-----Original Message-----
From: Chris Dagdigian [mailto:dag at sonsorol.org] 
Sent: Friday, January 20, 2006 9:54 PM
To: Kathleen
Subject: Re: [Bioclusters] Overcoming Grid drawbacks


The sales slant is coming on bit heavy even if it does seem you are actually
interested in the bioIT space.

A gentle word of etiquette for the biocluster list -- separate out your
questions and comments from mentions of your product and your employer.
There are many people active on bioclusters (and beowulf!) who work for
companies selling into the HPC space (myself included) but we all tend to
participate in these forums as individuals. It's not a rule or anything but
there is a definite culture that we've worked hard to maintain that
emphasizes discussions with a high content value and minimal (or no) sales
pitch.  Quite a few people in these forums post using personal email
addresses (as I do) so as to further maintain the noncommercial tone.


On Jan 20, 2006, at 9:59 PM, Kathleen wrote:

> Hey, can I take a poll?
> I'm learning that many enterprises are adverse to outsourcing BLAST, 
> for instance, even if it means freeing up maxed out resources.  I 
> haven't quite figured that out, but I'm sure someone will educate me.  
> Larry? :-) Furthermore, funding may not be such that acquiring 
> additional hardware resources is a viable "immediate" option.  Perhaps 
> that's why GRID computing is so attractive, but GRID computing comes 
> with it's own limitations.  What if enterprises could tap unused 
> resources on its network without the scheduling problems, data 
> collisions, and I/O problems?  What if users could propagate a 
> "grid-type" application automatically from one node and what if the 
> application could dynamical reassign resources on the fly when users 
> log on to nodes in use by the "virtual supercomputer"?  Is this of 
> interest in the bioinformatics community?
> Please educate me.  I really want to know what users need and want.  
> Thanks.
> Kathleen Erickson
> Massively Parallel Technologies, Inc.
> _______________________________________________
> Bioclusters maillist  -  Bioclusters at bioinformatics.org 
> https://bioinformatics.org/mailman/listinfo/bioclusters

More information about the Bioclusters mailing list