[OT?] Best Practices for Scientific Computing

Eray Ozkural (exa) erayo at cs.bilkent.edu.tr
Tue Aug 14 11:51:53 PDT 2001

Hash: SHA1

On Tuesday 14 August 2001 09:26 pm, Bob Drzyzgula wrote:
> Has anyone ever encountered a managment-oriented, MBA-speak
> document which outlines "best practices" for scientific and/or
> research computing? At work, we're struggling against an
> "enterprise-wide" deployment of Windows 2000 and Active
> Dirctory, the details of which [1] have us deeply concerned
> that we will soon be severely constrained in our ability to
> continue our use of Unix- and Linux-based systems as our
> primary computing platform. Research being only 10% of the
> organization, we're kind of in a tail-wagging-the-dog
> situaion, so the best-case outcome would probably be if
> we could simply convince them that they'll be better off
> just leaving us alone...

I am not an MBA-speak type of guy, and I don't claim to understand their aims 
in life but I do have a notion of telling technical stuff to non-technical 

  1. MBA people believe that they know a good deal about information systems 
(IMHO) so keep that in mind as you prepare a document. They use computers and 
I believe are very keen on using top-notch technology.

  2. They will like pretty pictures. For instance if you've used beowulf 
clusters, visualization of results/performance etc.

  3. They need to be convinced that UNIX is the proper technology for 
scientific work. Most of the tools they need are available on UNIX but not on 
windows. Example: beowulf software.

  4. Depending on 3, make the point that for scientific work the TCO is 
higher _and_ productivity of scientists will decrease.

  5. Network software is better in UNIX. Tell them why setting up UNIX 
servers and windows clients (for absolutely non-technical people) may be a 
good idea. Show them that there are great desktop environments on UNIX, 
demonstrate KDE and that in client terms windows won't be bringing any 
advantage. Also mention that UNIX provides better security: show that UNIX 
systems have never been affected by net worms/virii in the way windows boxes 
were recently.

- -- 
Eray Ozkural (exa) <erayo at cs.bilkent.edu.tr>
Comp. Sci. Dept., Bilkent University, Ankara
www: http://www.cs.bilkent.edu.tr/~erayo
GPG public key fingerprint: 360C 852F 88B0 A745 F31B  EA0F 7C07 AE16 874D 539C
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org


More information about the Beowulf mailing list