[Gllug] Lots of computing power required - is Amazon EC2 an answer?

[John Hearns]

On 30 June 2011 23:05, Adrian McMenamin <adrian at mcmen.demon.co.uk> wrote:
> I have struggled to find an algorithmic way out of this - but cannot see one.
>
> I have some code which is very parallelisable but requires huge computing
> resources - on a AMD 64 dual box I have here I reckon it will take about
> 20 days to run - but if I had access to about 600 CPUs it might take 20 -
> 30 minutes (it's run against a massive XML file - 800 MB or so - with
> different parameters)

As dsicussed, you could use Amazon EC2, or ask at your college for
access to a Condor pool.

Another alternative would be to build your own cluster.
You would not need to buy new machines - though if you did have the
budget, all you would need is motherboards, and just stakc them up
caseless and link them with a switch.

I should ask though - what language are you using?
Is it something which is amenable to running on a GPU? I ask as
Portland have recently announced compiler directives for GPUs,
and as a student you should be able to get a cheap or free license:

http://www.clustermonkey.net//content/view/305/1/
--
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug