itaps-parallel push-and-forget?

Carl Ollivier-Gooch cfog at mech.ubc.ca
Thu Nov 13 11:43:59 CST 2008


Tim Tautges wrote:
> Hi all,
>   Several functions are currently labeled as "push-and-forget".  Does 
> that mean we assume one-sided communication and processing, or that the 
> implementation has some mechanism for polling during unrelated function 
> calls?  What if the application doesn't call any iMeshP functions 
> besides one of this type?

My assumption is that if an app is calling one or more push-and-forget 
functions, that it also has to call a polling function to catch data 
that might have been pushed at it.  This is obviously different from the 
non-blocking send / blocking receive pair that solver apps typically 
use, but meshing apps can't always organize their communication this way 
(at least, it's not the most natural way to do it).

Carl

-- 
------------------------------------------------------------------------
Dr. Carl Ollivier-Gooch, P.Eng.                   Voice: +1-604-822-1854
Associate Professor                                 Fax: +1-604-822-2403
Department of Mechanical Engineering             email: cfog at mech.ubc.ca
University of British Columbia              http://www.mech.ubc.ca/~cfog
Vancouver, BC  V6T 1Z4                  http://tetra.mech.ubc.ca/ANSLab/
------------------------------------------------------------------------



More information about the itaps-parallel mailing list