MPI Tutorials: Getting started with LAM/MPI

1. INTRODUCTION

1.1 About this document

This document describes the steps to preparing an MPI session using LAM/MPI 7.1.4. We have organized the document into six sections:
1. Introduction

2. Preliminary setup (only needs to be executed once),

3. Compiling MPI programs,

4. Booting LAM/MPI,

5. Running under LAM/MPI, and

6. Shutting down LAM/MPI.

This document will not cover in details the ethical and proper computing etiquettes relating to the use of public clusters for parallel computing. As a result, users are requested to consider the affects of the improper use of LAM.

1.2 About MPI

MPI is suitable for parallel machines such as the IBM SP, SGI Origin, etc., but it also works well in clusters of workstations. Given the availability of clusters of workstations, using MPI to utilize them as a single parallel resource is an attractive solution.

The MPI-1 standard supports the portability and platform independent computing. As a result, users will undoubtedly enjoy the cross-platform development capability as well as heterogenous communication. For example, MPI codes which have been written on the RS-6000 architecture running AIX can be ported to a Sparc arhitecture running Solaris with almost no modification necessary.

1.3 About LAM

LAM is a daemon based implementation of MPI. Initially, the program lamboot spawns LAM daemons based on the list of host machines provided by the user. These daemons remains idle on the remote machines until they receive a message to load the MPI binary to begin execution. Bottom line:

• You boot LAM, it spawns daemons on your machines.
• You execute MPI programs while the LAM daemons exist in the background.
• Finally at the conclusion of the MPI session, you shutdown LAM.