[Bioclusters] Journalist request: linux cluster user feedback

[Alan Hart]

As below ...
Let me know if you want any more detail.
Alan Hart 
UNIX Manager 
Oxford GlycoSciences

> -----Original Message-----
> From:	Bernadette Toner [SMTP:btoner@genomeweb.com]
> Sent:	19 September 2002 17:44
> To:	bioclusters@bioinformatics.org
> Subject:	[Bioclusters] Journalist request: linux cluster user
> feedback
> 
> 
> Hi everybody -
> 
> I'm conducting an informal survey on user trends for bioinformatics Linux
> clusters for BioInform newsletter and I'm looking for as much input as
> possible.
> 
> The set of questions below has already been sent out to around 40 people
> who
> use Linux clusters for bioinformatics. If anyone on this list has a few
> minutes to add their two cents, that would be really helpful. Feel free to
> forward this e-mail to anyone else you know of who may be interested in
> participating.
> 
> I'll post the results to this list once the article comes out - probably
> in
> about two weeks.
> 
> Here's the questions (please respond directly to btoner@genomeweb.com):
> 
> - How long have you had your Linux cluster running?
> 
	1 year

> - How many nodes was your cluster originally? How many nodes is it now? Do
> you have plans to add more in the future?
> 
	10 now 20 expect to go 100+ in 2003

> - Did you build your cluster yourself or use an independent consultant or
> hardware vendor (If so, please specify who)?
> 
	Total self build (based on previous experience in the Oil Industry)

> - What type of processors make up your cluster?
> 
	Intel 1.2 GHz's

> - What type of distributed resource management software (ie, LSF, PBS, Sun
> Grid Engine) do you use?
> 
	Mosix Kernel level load sharing with Sun Grid queuing resource on
top

> - What bioinformatics/life science applications do you run on the cluster?
> Has this changed at all in the time you've had it running?
> 
	Sometimes daily changes !  Blast, fgenes, clustalw, emboss, fasta,
iprscan and lots more ...

> - Do you use a parallel Blast application? If so, which one (ie, BeoBlast,
> NBlast, Blackstone PowerBlast, TurboBlast, Paracel Blast)?
> 
	Nope.

> - Is the price/performance of your Linux cluster better or worse than you
> expected it to be? Please elaborate, if possible.
> 
	Good although my opinion is take more man hours to support than an
equivalent UNIX (e.g. solaris) cluster

> - Is the total cost of ownership of your Linux cluster (including air
> conditioning/cooling, sys-admin costs, downtime, footprint, etc.) better
> or
> worse than you expected it to be? Please elaborate, if possible.
> 
	Good.

> - Overall, what are the biggest advantages to using a Linux cluster in
> bioinformatics? What are the biggest drawbacks?
> 
	Opensource flexible build, low cost, easy to modify, good price
performance, lots of open source BioInf. code, dual boot NT & linux for more
flexibility, there is nothing else out there like "mosix" !

> - What future improvements would be on your wish list for Linux cluster
> technology for bioinformatics applications?
> 
	More snmp c.f. mosix does CPU level load sharing, often apps. are
memory bound.
	More processors per nodes or kernel level snmp

> All responses will be treated as confidential. If you're willing and have
> the time to discuss any of these questions further, please let me know
> what
> day and time would be most convenient for a follow-up discussion.
> 
	Not willing to discuss but interested in other responses.
	OGS config is confidential please.

> Thanks in advance to all -
> 
> Bernadette Toner
> 
> ......................................
> Bernadette Toner
> Editor, BioInform
> GenomeWeb LLC
> 212-651-5636
> btoner@genomeweb.com
> www.bioinform.com
> 
> 
> 
> _______________________________________________
> Bioclusters maillist  -  Bioclusters@bioinformatics.org
> https://bioinformatics.org/mailman/listinfo/bioclusters


********************************************************************
The information transmitted by this email is private and
confidential and is intended for the use of the intended 
recipients specified therein.
If you are neither an intended recipient nor an employee 
or agent responsible for delivery to an intended recipient, 
you should be aware that any dissemination, distribution 
or copying of this communication is strictly prohibited.
If you received this communication in error, please 
notify us immediately.
********************************************************************