MPI - Introduction

You will learn about scientific parallel programming on the AVIDD Linux Cluster at Indiana University.

You will use the LAM Message Passing Interface (MPI) library to write, compile, and run a set of example parallel programs. Workshop examples will be in the 'C' language.
- Another point to note: As mentioned above, we will illustrate execution of example programs using LAM. If you would prefer to use (or are interested in learning about use of) MPICH, then check out the Compiling and Running Parallel jobs using MPICH on AVIDD section.
You will trace and profile your parallel programs with Intel Trace Collector/Trace Analyzer. This will help you analyze your code to achieve optimal performance.
You will debug parallel programs using the interactive TotalView debugger.
Caveat: You will not attain expertise from this tutorial, but it should get you started!