Are we going mainstream?

[Schilling, Richard]

Nice to hear this Gerry. 

I think we are going mainstream.  And because of that, this is a good time
as any to openly discuss how to STAY mainstream.  I fear the day when my
supervisor says he prefers the "Microsoft clustering solution" over anything
the Beowulf movement has produced. You just know it's comming.  Yech.

Here are some things our movement should consider:
	* Marketing in general; counter-advertising to competing
technologies as well.
	* Approval of standards; evaluation of software products
	* Benchmarking
	* Continued research (technology and market)
	* Sponsorship of research, educational, and public awareness
programs

	we do a lot of this already, it's just not well publicized, which is
why I put marketing (which includes marketing research) at the top of the
list.

Here's Microsoft's white paper on the subject:
http://www.microsoft.com/ntserver/ntserverenterprise/techdetails/prodarch/cl
ustarchit.asp

This deserves a systematic tearing apart, which I'll be doing as time
permits.

And since I brought it up, might as well give an opinion: as companies
continue to market Beowulf technologies, we stand to be moved into obscurity
as some of the early adopters.  For example Microsoft's adoption of newer
clustering technology (e.g. Beowulf) would not do the movement any good, and
in fact might damage it. They tend to embrase, extend, and eliminate
percieved rivals to their products and licencing policies.  Movements like
the Beowulf movement do more good than any single company can hope to do.

Richard Schilling
Webmaster/Web Integration Programmer
Affiliated Health Services
Mount Vernon, WA

http://www.affiliatedhealth.org
my other web site: http://www.nationalinformatics.com





> -----Original Message-----
> From: Gerry Creager N5JXS [mailto:gerry at cs.tamu.edu]
> Sent: Friday, November 17, 2000 6:07 AM
> To: beowulf at beowulf.org
> Subject: Are we going mainstream?
> 
> 
> I just got back from a meeting of the Great Plains Network in Kansas
> City, where discussion of Beowulfery was rampant.  And casual.  One
> participant mentioned the several clusters they have at North Dakota. 
> And the little one he keeps for personal projects in his 
> office.  Okay,
> okay, so several (or most) of us have /Wulfs in our offices, still it
> was a little interesting to hear about 'em from folks I don't see on
> this list, mentioned in such casual way.  
> 
> NCSA's implementing a general access 'Wulf.  UCSD has at 
> least one, and
> is implementing more.  Classes in PVMN and MPI are commonplace.
> 
> Another issue discussed was the use of what I can only call now
> "distributed beowulfery" in the general form of, say, seti at home, was
> discussed not so much as something to look into, but something to
> implement in the Internet 2 community.
> 
> All in all, an interesting meeting.
> --
> Gerry Creager -- gerry at cs.tamu.edu
> Network Engineering			|Research focusing on
> Academy for Advanced Telecommunications |Satellite Geodesy and 
> and Learning Technologies		|Geodetic Control
> Texas A&M University	979.458.4020  (Phone) -- 979.847.8578  (Fax)
> 
> _______________________________________________
> Beowulf mailing list
> Beowulf at beowulf.org
> http://www.beowulf.org/mailman/listinfo/beowulf
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.scyld.com/pipermail/beowulf/attachments/20001117/b0c94355/attachment.html