Hi<br><br>Analyzing performance of tcp and lamd rpi on a Linux cluster (Linux Kernel 2.4.21-52), I&#39;ve found that for an MPI application with (only) collective communications completion of the application got slowed down with tcp rpi, but not with lamd rpi when they were run on a set of busy nodes. It is somewhat strange because what I understood from User&#39;s manual and the design documentation of RPI SSI module (1.0-ssi-rpi.pdf) was that the tcp and lamd rpi apply to point-to-point communications, not to collective communications. Is it misunderstanding? <br>
<br>So... My questions are:<br>1) does tcp rpi module and lamd rpi module work differently for collective communications like MPI_Reduce?<br>and <br>2) is there a more detailed specification, manual, or reference for tcp and lamd rpi implementation?<br>
<br>Any helpful comment would be appreciated.<br><br>Jeho<br>