[gmx-users] Apples-to-Apples comparison for dpcc benchmark

[Anton Feenstra]

David wrote:
>>Are there any
>>other optimization options I should be aware of when making a comparison?
>>
> 
> This is all.

Well, as Berk hinted, there will be drastic changes in the upcoming new 
version, like domain composition, threading. Sorry to say I don't know 
more details about it.

Additionally, but these you must surely be aware of already, there are 
of course the compiler, compiler options, and of course the (mpi) 
communication. In the commonly used lam-mpi or mpi-ch, there are a 
number of parameters that you can tune, mainly to minimize latency 
(bandwidth is generally not an issue).


-- 
Groetjes,

Anton

* NOTE: New Phone & Fax numbers (below) *

  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 59 87608 - Fax: +31 20 59 87610           |
|             | Feenstra at few.vu.nl - www.few.vu.nl/~feenstra/         |
|             | "If You See Me Getting High, Knock Me Down" (RHCP)    |
|_____________|_______________________________________________________|