Newsgroups: comp.parallel.pvm
From: schiotz@oersted (Jakob Schiotz)
Reply-To: schiotz@fysik.dtu.dk
Subject: Losing messages on CM-5E
Organization: Physics Department, Techn. Univ. of Denmark
Date: 24 Oct 1994 17:58:27 GMT
Message-ID: <38gsk3$svv@news.uni-c.dk>

Hi,

I am running a PVM application on a 128 node CM-5E. The processors are
configured as a 3D grid, communication is very simple: Each processor
receives a message from the host program, and retransmits it to the 26
neighbours. Some time later this is repeated, again and
again. However, after some time a message is lost. It has been sent,
but is never received. If I rerun the application, it will happen at
another time between who other nodes. There is nothing in the
/tmp/pvml.XXX file. Has anybody seen this, or am I just unlucky?

BTW, the grid is only two wide in two directions with my test file, so
some of the messages are exact copies. Can this have an influence?

Any ideas?

--
Jakob Schiotz                    !  The noble art of losing face
Physics Department               !  May someday save the human race
Technical University of Denmark  !  And turn into eternal merit
DK-2800 Lyngby, Denmark          !  What weaker mind would call disgrace
schiotz@fysik.dtu.dk             !                  - Piet Hein
              ^^^ Note: new domain

