[Beowulf] Is Beowulf a pure compute only cluster

[Robert G. Brown]

On Mon, 11 Apr 2005, Richard Chang wrote:

> I just want to clarify if "Beowulf" is a pure compute only cluster. By
> this I mean if it is used only for number crunching. Or is it can be
> used as Fail-Over cluster in production environments like DATA-CENTERS
> where the data-uptime is of utmost importance. Can we use Beowulf for
> Mission critical database environments like SAP or Oracle
> implementations also.

These questions are answered in a variety of documents e.g. the beowulf
FAQ, my online book, the beowulf HOWTO and more.  In a nutshell,
"Cluster" is a broad term that includes high availability (HA) clusters
(failover servers, data centers) and high performance computing (HPC,
aka "supercomputing") clusters.  Beowulf is a term for a specific
configuration (single headed, passive node) HPC cluster, where there are
other HPC cluster configurations such as NOW or Grid or compute farm
that are equally or more common (but less useful for fine grained
parallel computations).

Further details in many documents that Google will turn up.

> What are the exact uses of a Beowulf.

Manifold fine-grained parallel computations in science and engineering,
too many to list here.  Also visualization and movie making.  Probably
still more things.

 HTH

   rgb

> 
> RC

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu