Newsgroups: comp.parallel.pvm
From: link@enif.astro.indiana.edu (Robert!)
Subject: question on pvm_reduce
Organization: Indiana University
Date: 9 Feb 1996 16:04:32 GMT
Message-ID: <4ffrag$gq7@usenet.ucs.indiana.edu>

When using pvm_reduce, when and where does the actual computation of
the global function (e.g. PvmSum) occur.  Specifically, if I call a
pvm_reduce for a group, followed by a pvm_barrier for the same group,
am I guaranteed that by the time the root leaves the barrier the
reduced data is in place?  If not, how do I determine when the
reduction is complete?


-- 
-r	| ``What lies behind us and what lies before us are
	|   tiny matters compared to what lies within us.''
	| 		--Ralph Waldo Emerson


