Literature Survey on a New Advanced Refactoring Based Approach for Parallelism Using Heterogeneous Parallel Architectures

Shanthi Makka, Bharat Bushan Sagar

Abstract: Refactoring is the process of changing the structure of a program without changing its behavior. Refactoring has so far only really been deployed effectively for sequential programs. However; with the increased availability of multi core systems; refactoring can play an important role in helping both expert and non-expert parallel programmers structure and implement their parallel programs. This paper describes benefits or advantages of a refactoring approach for parallel programs using heterogeneous parallel architectures such as GPUs and CPUs. A refactoring based methodology gives many advantages over unaided parallel programming: it helps identify general patterns of parallelism; it guides the programmers through the process of refining a parallel program; whether new or existing; it enforces separation of concerns between application programmers and system programmers; and it reduces time to deployment. All of these advantages help programmers understand how to write parallel programs.

Keywords: refactoring, parallelism, CPU, GPU, refactoring tool