__/ [ TheLetterK ] on Thursday 18 January 2007 04:07 \__
> mlw wrote:
>> I saw someone mention "Windows Compute Cluster Server" on this group and I
>> had to take a look.
>>
>> OK, is it me or is it a stupid attempt at a product? At its core, it is
>> based on MPI, specifically, "Argonne National Labs Open Source MPI2
>> implementation."
>>
>> Here are my problems with it:
>> It is limited to a 64bit implementation.
>> It seems to REQUIRE, based on their overview (I could be wrong) a
>> dedicated cluster topography.
>> They are trying an "embrace, extend, extinguish" on MPI with "MS-MPI."
>>
>>
>> Multiple implementations of MPI and PVM are available for Linux, what is
>> the "value add" of "Windows Compute Cluster Server?"
>>
>
> What would displace Linux as the king of clusters? Is Microsoft going to
> stop charging by the node?
I think that, at the moment, their product is so undesirable, that they are
sometimes willing to give it for free. Those who accept it become part of an
expriment (see my earlier post).
--
~~ Best regards
Roy S. Schestowitz | Othello for Win32/Linux: http://othellomaster.com
http://Schestowitz.com | RHAT Linux | PGP-Key: 0x74572E8E
8:05am up 91 days 18:19, 7 users, load average: 0.35, 0.31, 0.27
http://iuron.com - Open Source knowledge engine project
|
|
