To Parallelize or Not to Parallelize, Speed Up Issue

TL;DR

This paper discusses the decision-making process for parallelizing applications using MPI, analyzing when parallelization is beneficial and proposing an experimental method to predict potential speedup.

Contribution

It introduces an experimental approach to predict MPI application speedup, aiding in deciding whether parallelization will be advantageous.

Findings

Parallelization benefits depend on balancing overhead and speedup

Proposed method helps predict MPI application performance

Guidelines for deciding on parallelization are provided

Abstract

Running parallel applications requires special and expensive processing resources to obtain the required results within a reasonable time. Before parallelizing serial applications, some analysis is recommended to be carried out to decide whether it will benefit from parallelization or not. In this paper we discuss the issue of speed up gained from parallelization using Message Passing Interface (MPI) to compromise between the overhead of parallelization cost and the gained parallel speed up. We also propose an experimental method to predict the speed up of MPI applications.