re: help

[Greg Freemyer <freemyer@NorcrossGroup.com>]

 >>  Hi,
 >>  cd u guide me how to learn abt clustering in linux.

 >>  Thanks
 >>  Regds
 >>  Manish


 >>  --
 >>  Things work out best for those who make the best use of it.


There are at least a dozen ways to cluster Linux boxes with totally different goals in mind.

What kind of clustering do you want to do?  A few possibilities are:

1) Director Based (usually used for static web content)    (i.e. LVS)

2) Fail-over between 2 nodes.   (i.e. I think Heartbeat, Failsafe, Kimberlite, and Mission Critical Linux all do this.) 

3) Active-Active of a small number of nodes.   (i.e. SSI-Linux (alpha quality code only currently), maybe others)

4) Performance clusters   (i.e. Beowulf and others, I know very little about this)

Once you say what is you want to do, maybe someone can point you towards a reasonable technology.

BTW: about 18 months ago I started to put some background stuff on our companies website you might find useful.  If is definitely preliminary, and at this point it is getting dated, but it might be worth scanning:  http://www.norcrossgroup.com/Documents/Clusters/types.htm

Greg Freemyer
Internet Engineer
Deployment and Integration Specialist
The Norcross Group
www.NorcrossGroup.com

-
Linux-cluster: generic cluster infrastructure for Linux
Archive:       http://mail.nl.linux.org/linux-cluster/