An efficient processor allocation scheme for mesh connected parallel computers

Several processor allocation schemes are proposed in the literature for mesh connected parallel computers. All these schemes aim at improving the system performance by reducing internal fragmentation or by enhancing the submesh recognition ability. In this paper, we propose an approach of system partitioning to reduce external fragmentation and thereby improve the system performance. The target systems considered here are two-dimensional meshes where the side lengths are powers of two. Processors are allocated to a partitioned mesh based on their submesh size requirements. The proposed scheme can be implemented in conjuaction with any of the existing 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 utilazation are obtained. It is observed that, in most cases, the proposed scheme demonstrates better performance. Time complexity of the proposed scheme is less by a factor of n compared to the corresponding allocation scheme without partitioning, where n=log/sub 2.