On Sat, Apr 14, 2012, Albert wrote:
>
> I am going to use Amber for hundreds of ns of a 40,000 atoms system.
> Does any body has any idea how many core should I use the most so that
> the efficiency can be the best? suppose each nodes has two sockets, each
> socket with 6 cors and 12 cores per modes.... Will 504 cores be a good
> choice?
These are questions you need to just find out for yourself, with your system
on your machine. Run a few short test calculations before embarking on
simulations of hundreds of nanoseconds. We do provide lots of benchmarks on
the web site, which should give a little help. But every machine is
different, and there is no substitute for having your own data.
Having said all that 500 cores is almost certainly way too many. What number
is best depends mostly on the nature of the interconnect between nodes.
...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Apr 14 2012 - 07:00:04 PDT
