Clustering, databaze a sdileni dat

[Jiri Valenta]

On Sun, Dec 12, 1999 at 06:40:26PM +0100, Michal Krause wrote:
> Zdravim,
> 
> jsem (pro zmenu) zase trosku OT, ale nemam se kde jinde zeptat.
> Premyslim docela vazne o nejake paralelizaci Navrcholu, protoze jeden
> stroj zvolan prestava stihat a uz neni kam zvedat vykon (vyjma SMP, u
> ktereho mi ale vadi, ze za nejaky cas zase nebude co zlepsit, v tomto
> ohledu mi pripada clustering pruznejsi).

Osobni zkusenosti nemam, ale kdysi jsem si schoval z linux-kernel tohle: (snad to trochu pomuze). V posledni dobe taky neco probehlo, ale to nemuzu najit (snad az bude chvilka).



Jeremy Y.Y.Lai wrote:
> 
> Anyone has suggestion of where to start, what to watch out etc.. if I want
> to do preemptive process migration among a cluster of workstations running
> Linux.

simple question : do you want to add support for process migration to
linux
or do you want to use several workstations to do parallel computing ?

if you answer the second choice then take a look at these :

http://www.aa.net/~mtp/index.html
http://yara.ecn.purdue.edu/~pplinux/pphowto.html
http://yara.ecn.purdue.edu/~pplinux/ppcluster.html
http://www.cs.cornell.edu/Info/People/mdw/hpc/hpc.html
http://cesdis.gsfc.nasa.gov/linux/beowulf/beowulf.html

There may be double/these pages will likely contain links to each other.
(and I haven't checked these for a long time, but it should work)

  hope it helps

  alex

-- 
\\-----------------------------------------------------------\\
// Alexandre Maret -- Linux : The choice of a GNU generation //
\\ amaret na infomaniak.ch -- http://www.infomaniak.ch/~amaret  \\
//-----------------------------------------------------------//