On Sat, 26 Jun 2010, Edward Ned Harvey wrote:

> I am aware of this.  I've been a long time user and supporter of dells and
> custom built blackboxes, workstations and servers, running usually linux of
> various kinds, clustered via SGE.

Hi Edward.  I misunderstood sorry.  I read your original email as 'I'm 
getting in to HPC' and responded as such :)

> The root concept is:  Break away from the assumption of xeon or equivalent
> amd processors.  Jump down to the super small, super low-power, super cheap
> class of processors, atom, arm, etc... and use them to beat the xeons for
> some situations, such as distributed work load, or server virtualization.

I'm quite open to this possibility.  One problem with using many small 
processors is that many problems _are not_ heavily parallelizable[1].  If 
you can break your task into thousands of subtasks then great - it may 
well be an good choice but there are alot of variables involved.

[1] Waaaay back at University I got to code on a MASPAR[2] for several 
courses, which was great fun.  4096 CPUs in a 64x64 array.  Suffice to say 
the programming assignments did involve highly parallelizable tasks like 
heat flow over a surface :)

[2] http://en.wikipedia.org/wiki/Maspar

Cheers,

Rob

-- 
Email: [email protected]         Linux counter ID #16440
IRC: Solver (OFTC & Freenode)
Web: http://www.practicalsysadmin.com
Open Source: The revolution that silently changed the world
_______________________________________________
Tech mailing list
[email protected]
http://lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to