An Approach to Parallel Processing

Abstract

Parallel processing offers enhanced speed of execution to the user and facilitated by different approaches like data parallelism and control parallelism. Graphic Processing Units provide faster execution due to dedicated hardware and tools. This paper presents two popular approaches and techniques for distributed computing and GPU computing, to assist a novice in parallel computing technique. The paper discusses environment needs to be setup for both the above approaches and as a case study demonstrate matrix multiplication algorithm using SIMD architecture.