Design of Image Processing Algorithms for Multicore Processor: Implementation and Performance Analysis on Cell Broadband Engine

Abstract

Enhancement of CPU processing power and speed has been a priority issue in the past several years. Conventional approaches to satisfy this demand were to improve the performance of the core singularly. However the CPU design became much more complex, causing problems as increases in power consumption, over-heating and manufacturing cost. That’s why world is moving towards multi-core configurations. Multi-core architecture offers higher performance at same power consumption and manufacturing cost.

The Cell Broadband Engine Architecture (CBEA) is a multi-core architecture, including a 64-bit PowerPC Processor Element and eight Synergistic Processor Elements which offers a raw compute power of up to 200 GFlops per 3.2 GHz chip. The Cell bears a huge potential for compute intensive applications like Image processing applications, scientific applications etc. Image processing applications such as filtering, edge detection, correlation etc. require large computation power. Processing large images on a single core system consumes a lot of time. Such applications can be ported on CBEA and performance can be improved. However it requires addressing many challenges for implementing such applications for CBEA.

The goal of this dissertation is to implement image processing algorithms for CELL Broadband Engine, address the challenges occurring because of multiple cores and achieve higher performance than conventional systems.