Newsgroups: comp.parallel.mpi
From: calvin@drus.alpes.cea.fr (Calvin_Christophe)
Subject: Re: Q:  Performance of collective communications?
Organization: CEA-Commisariat a l'energie atomique
Date: 26 Sep 1996 09:47:53 GMT
Message-ID: <52djg9$9sg@news.cea.fr>

In article <52brdk$jmm@news.cc.utah.edu>, Michael Pernice <usimap> writes:
> Take a look at IEEE Parallel & Distributed Technology, (4)1, Spring 1996.
> It contains an article by Xu and Hwang that models communication on a
> large IBM SP2.  It is one of the few 'communication performance evaluation'
> articles I've seen that evaluates collective as well as point-to-point
> communication operations.
> 
> BTW that article shows poor latency characteristics for gather/scatter and
> total exchange for the MPICH implementation.  Is this being addressed?
> -- 
> _______________________________________________________________________________
> 
> Michael Pernice                          Center for High Performance Computing 
> Phone:  (801) 581-7708                   85 Student Services Building  
> Fax:    (801) 585-5366                   University of Utah
> E-mail: usimap@sneffels.usi.utah.edu     Salt Lake City, Utah 84112
> _______________________________________________________________________________
> 

You can find also some informations about performances of collective
and point-to-point communications on Cray T3D using PVM and Shmem
in the following paper :
"Performance Evaluation and Modeling of Collective Communications on Cray T3D"
by C. Calvin and L. Colombet.
This paper has been accepted for publication in the Parallel Computing Journal.



    	Christophe CALVIN     				  
+-------------------------------------------------------+
| calvin@alpes.cea.fr        	DRN/DTP/STR/LMTL, 	|
| Christophe.Calvin@imag.fr  	CEA Grenoble, 		|
| tel: (33) 76.88.32.04  	17 avenue des Martyrs, 	|
| fax: (33) 76.88.94.53         38054 Grenoble cedex 9,	|
| 			  	FRANCE		 	|
+-------------------------------------------------------+




