SCIENTIFIC-LINUX-USERS Archives

December 2007

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jan Iven <[log in to unmask]>
Reply To:
Date:
Wed, 12 Dec 2007 09:06:18 +0100
Content-Type:
text/plain
Parts/Attachments:
text/plain (39 lines)
On 12/12/07 01:48, Fernando C. Estrada wrote:
> El mié, 12-12-2007 a las 10:33 +1000, Michael Mansour escribió:
..
>> There are so many which all serve different purposes. What purpose are you
>> trying to serve?
>>
>> Michael.
> 
> It's a basic cluster, 12 computers (i386), and the purpose is learn
> about instalation and some of administration. I have experience with
> Debian, but I don't know if I need special learning about SL or if only
> with manuals about RedHat.

Red Hat manuals should be fine for SL or CentOS. Debian keeps some 
config/system files in different places, and has a different installer.

 > The goal is run scientific applications, and
> integrate with the Grid team in the University, so, I think that I need
> high processing clusters knowledge too.

In which case you probably should read up on

1. Kickstart installations (Google, T.U.V has 
http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/sysadmin-guide/ch-kickstart2.html 
)

2. batch system (ask you Grid guys for their preferred system -
   PBS/Torque/CONDOR/...) - so you actually can "integrate".

3. a way to maintain the machines. "SSH" may do for the beginning, but 
to ensure long-term consistency you may want to have a look at automated 
configuration management (cfengine/puppet/quattor/..).

An alternative that combines 1. and 3. is NCAPI "ROCKS", 
http://www.rocksclusters.org

Regards
Jan

ATOM RSS1 RSS2