From: Yon Han CHONG Newsgroups: comp.parallel.mpi Subject: Re: LAM vs. MPICH Date: Mon, 16 Nov 1998 14:16:54 +0100 Organization: Cranfield University Message-Id: <365025C6.16C04592@cranfield.ac.uk> References: <364C00E2.2876EEBD@cranfield.ac.uk> <364C895A.40E9D80B@scl.ameslab.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=EUC-KR Content-Transfer-Encoding: 7bit Vasiliu Bogdan wrote: > > Yon Han CHONG wrote: > > > What's your option on LAM and MPICH. I have seen posts to this newsgroup > > talking about mixing and match LAM and MPICH. What are the advantages? I > > am especially interested in performance on a Linux dluster but any > > comment will be welcomed. > > > > -- > > ---------------------------------------------------------------------- > > Yon Han Chong | > > Flow Control and Prediction Group | Tel: + 44 1234 750 111 ext. 5420 > > Cranfield College of Aeronautics | E-mail: Y.H.Chong@cranfield.ac.uk > > Cranfield University | Fax: + 44 1234 752 149 > > United Kingdom | Mobile: + 44 958 381 395 > > ---------------------------------------------------------------------- > > Yon. > > I am using MPICH on a cluster of PCs running Linux. My performance tests > (ping-pong) > showed that LAM had a very low bandwidth compared to MPICH or PVM. On a > Fast > Ethernet (100Mbits/sec) connection LAM achieved 30 Mbits/second while > MPICH and PVM achieved > 70 Mbits/second. I have a graph that shows the bandwidth curves, if you are > interested I can send the '.ps' > file to you. > Yes, please. -- ---------------------------------------------------------------------- Yon Han Chong | Flow Control and Prediction Group | Tel: + 44 1234 750 111 ext. 5420 Cranfield College of Aeronautics | E-mail: Y.H.Chong@cranfield.ac.uk Cranfield University | Fax: + 44 1234 752 149 United Kingdom | Mobile: + 44 958 381 395 ----------------------------------------------------------------------