Processor Allocation Using Partitioning in Mesh Connected Parallel Computers

Prasant Mohapatra

Abstract: Several processor allocation schemes for mesh connected parallel computers have been proposed in the literature. These schemes aim at improving system performance by reducing internal fragmentation or by enhancing submesh recognition ability. In this paper, we propose a system partitioning approach to reduce external fragmentation and thereby improve system performance. The target systems considered here are two-dimensional meshes where the side lengths are powers of 2. Processors are allocated to a partitioned mesh based on their submesh size requirements. The proposed scheme can be implemented in conjunction with any of the existing processor allocation schemes and thereby can also exploit the advantages offered by those schemes. The performance measurements are done through simulation experiments. Completion time for a fixed number of jobs, internal and external fragmentation, and system utilization are measured as performance indicators. It is observed that, in most cases, the proposed scheme demonstrates better performance than the previously proposed algorithms. Time complexity of the proposed scheme is less by a factor ofnthan the corresponding allocation scheme without partitioning, wheren= log2{min(w,h)}, andwandhare the width and height of a two-dimensional mesh.