[Bioclusters] Cluster ideas/suggestions

[Nick D'Angelo]

Thank you all, I guess I have some homework/setup to do.

Thanks again,

Nickd

-----Original Message-----
From: Joe Landman [mailto:landman at scalableinformatics.com]
Sent: Thursday, January 13, 2005 1:02 PM
To: Clustering, compute farming & distributed computing in life science
informatics
Subject: Re: [Bioclusters] Cluster ideas/suggestions


FWIW:  you can pick up Fedora compatible RPMs for ClustalW and NCBI 
BLAST (and HMMer) at http://downloads.scalableinformatics.com .

Micha Bayer wrote:
> On Thu, 2005-01-13 at 15:54, Nick D'Angelo wrote:
> 
>>Thank you Micha.
>>
>>Just so I am clear on this, do I need to migrate to Biobrew or can I run
>>Condor on my current three machines (FC2) to achieve the same thing - a
>>cluster of machines running blast queries and Clusa1W?
> 
> 
> No, you should be okay just using Fedora - we do. You just have to get
> your bio apps separately (e.g. from NCBI or EBI) and then you can just
> preinstall them on all your machines - that saves even more time.
> 
> Condor installation takes minutes and it works out of the box usually.
> It's a great example of how good open software can be if it has matured
> over a number of years (since 1989 in this case).
> 
> If you want I can give you our config files - that would save you time.
> With our config files we force our jobs to run always (regardless of
> user activity) because our cluster is pretty much dedicated to running
> jobs at the moment, with only the occasional unsuspecting student
> wandering in.... :-)
> 
> cheers
> Micha
> 
> 
> 
>>Sorry for the repeated questions, I guess I venturing into unknown (to me)
>>territory.
>>
>>Nick
>>
>>
>>
>>-----Original Message-----
>>From: Micha Bayer [mailto:michab at dcs.gla.ac.uk]
>>Sent: Thursday, January 13, 2005 10:44 AM
>>To: Clustering, compute " farming & distributed computing in life
>>science informatics
>>Subject: RE: [Bioclusters] Cluster ideas/suggestions
>>
>>
>>Nick,
>>
>>if you have dedicated machines then Condor and BLAST can work reasonably
>>well together because you can keep target databases preinstalled on all
>>your machines. This gets you around the i/o issues.
>>
>>We run BLAST on Condor that way and it seems pretty reasonable. There is
>>also a very large Condor Pool at Wisconsin Uni which is used for BLAST
>>and it does a hell of a lot of useful work. 
>>
>>Micha
>>
>>
>>On Thu, 2005-01-13 at 15:34, Nick D'Angelo wrote:
>>
>>>All great ideas, thank you for them so far.
>>>
>>>I should have mentioned, that the cluster is required to run Blast,
>>
>>CLusta1W
>>
>>>and RNAfold.
>>>
>>>Currently I have three Fedora machines, kind of all doing independent
work
>>>in isolation of each other and we manually setup the jobs to run until
>>>completed.
>>>
>>>Nick 
>>>
>>>-----Original Message-----
>>>From: Micha Bayer [mailto:michab at dcs.gla.ac.uk]
>>>Sent: Thursday, January 13, 2005 10:01 AM
>>>To: Clustering, compute " farming & distributed computing in life
>>>science informatics
>>>Subject: Re: [Bioclusters] Cluster ideas/suggestions
>>>
>>>
>>>Hi,
>>>
>>>this sounds like you might want to look at Condor. Condor is an
>>>excellent tool for doing just that sort of thing (desktop cycle
>>>harvesting). 
>>>
>>>It can be installed on Linux or Windows machines, and it is highly
>>>configurable. One can set machines up so that jobs never run when the
>>>user is working, so that users don't even know a job is lurking in the
>>>background on their machine.  
>>>
>>>This would save you the trouble of having to reboot the machines, and
>>>you would get extra cycles during the day when users are in their coffee
>>>break etc. 
>>>
>>>Very nice package, free , very mature and great documentation. We run a
>>>Condor pool here and we think it is excellent. There is also a large
>>>user community already with mailing lists etc and good support
>>>otherwise.
>>>
>>>http://www.cs.wisc.edu/condor/
>>>
>>>If you have a pool of users who trust you to run jobs on their machines
>>>then this is just the ticket.
>>>
>>>cheers
>>>Micha
>>>
>>>
>>>On Thu, 2005-01-13 at 14:34, Nick D'Angelo wrote:
>>>
>>>>In speaking to our R and D managers, they posed a good question.
>>>>
>>>>What about their desktop machines (relatively New Dell PCs), have them
>>>>automatically shutdown and reboot, say at 8:00 pm and then join the
>>>
>>>cluster
>>>
>>>>until 05:00 am.
>>>>
>>>>Then at 5:00 am, initiate another reboot that will present the Windows
>>>
>>>Login
>>>
>>>>for the 'normal business day'.
>>>>
>>>>This would make great use of the hundreds of computers that are mostly
>>>>sitting idle during the off hours 'normal business day'.
>>>>
>>>>Thoughts, suggestions always appreciated, thanks.
>>>>
>>>>Nick
>>>>
>>>>In case you need to know,  I am likely going to install the BioBrew v3.x
>>>>which has just been pre-leased in Beta I believe.
>>>>_______________________________________________
>>>>Bioclusters maillist  -  Bioclusters at bioinformatics.org
>>>>https://bioinformatics.org/mailman/listinfo/bioclusters
> 
> 
> _______________________________________________
> Bioclusters maillist  -  Bioclusters at bioinformatics.org
> https://bioinformatics.org/mailman/listinfo/bioclusters

-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
phone: +1 734 612 4615
_______________________________________________
Bioclusters maillist  -  Bioclusters at bioinformatics.org
https://bioinformatics.org/mailman/listinfo/bioclusters