Interactive message debugger for parallel message passing programs using Lam-Mpi

Many complex and computation intensive problems can be solved efficiently using parallel programs on a network of processors. One of the most widely used software platforms for such cluster computing is LAM-MPI. To aid develop robust parallel programs using LAM-MPI we need efficient debugging tools. The challenges in debugging parallel programs are unique and different from those of sequential programs. Unfortunately available parallel debuggers do not address these challenges adequately; This thesis introduces IDLI, a parallel message debugger for LAM-MPI, designed on the concepts of multi-level debugging. IDLI provides a new paradigm for distributed debugging while avoiding many of the pitfalls of present tools of its genre. Through its powerful yet customizable query mechanism, adequate data abstraction, granularity, user-friendly interface, and a fast novel technique to simultaneously replay and sequentially debug one or more processes from a distributed application, IDLI provides an effective environment for debugging parallel LAM-MPI programs