[Beowulf] Please help to setup Beowulf

Geoff Jacobs gdjacobs at gmail.com
Tue Feb 17 17:28:17 EST 2009


Nifty Tom Mitchell wrote:
> On Tue, Feb 17, 2009 at 05:56:18AM -0600, Geoff Jacobs wrote:
>> Nifty Tom Mitchell wrote:
>>> On Mon, Feb 16, 2009 at 05:06:17PM +0530, Indrajit Deb wrote:
>>>>    Hello,
>>>>    I want to setup beowulf for my eight node cpu to run simulation. Debian
>>>>    4.0 is installed in each cpu. I am not familiar with beowulf. If any
>>>>    one help me to setup beowulf for the begining to the end i will be
>>>>    grateful.
>>> Hmm Molecular Biology and Bioinformatics....
>>> Which Molucular modeling code do you plan to run.
>>> Most but not all are built on MPI libraries does
>>> Debian have a pre-packaged MPI development and
>>> run time environment.  Look for things like Open MPI...
>> Yes, it does. OpenMPI, LAM, MPICH (but only version 1.2.x).
>>
>>> The simulation codes you plan to run will decide 
>>> a lot of things.  For example it may cause you to move from
>>> Debian  to ROCKS/Centos or something else.  Based
>>> on your question you are a Bio guy more than 
>>> a computer guy.  If so let your application codes drive
>>> the cluster decisions.
>> Why do you say this? Debian (5.0, what I checked just now) includes all
>> the basic Beowulf elements and even tosses in GROMACS pre-compiled for
>> OpenMPI. If it's what his people are used to there's no reason to switch.
>>
>> I would recommend using Debian 5.0 though, just so the latest GNU
>> compilers are available.
>>
> ....
> "> Why do you say this? ...."
> 
> Because the original poster did not disclose what he or his people
> have experience with.  It almost sounds as if he is off by himself.
> 
> We just do not know... 

OK, I think we've outlined the criteria by which he should make his
distro choice. I was assuming there was a reason he specified Debian
Etch in the OP.

-- 
Geoffrey D. Jacobs
_______________________________________________
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



More information about the Beowulf mailing list