All shapes busy list contiguous allocation strategy for 3D mesh multicomputers

Although contiguous allocation is known to suffer from high processor fragmentation, it is useful for security and accounting reasons. This is because the allocated jobs are isolated from one another, where each job is allocated to an exclusive submesh of processors, and the allocated sub-mesh has the same size and shape of the requested job. The size and shape constraint leads to high processor fragmentation. Most recent contiguous allocation strategies suggested for 3D mesh multiconputers attemp all possible rotations of the job request when allocation fails for the original job request, which reduces processor fragmentation and hence improves system performance. To generalize this restricted rotation, we propose in this paper a new contiguous allocation strategy for 3D mesh multicomputers, referred to as All Shapes Busy List (ASBL for short), which considers all possible contiguous request shapes when attempting allocation for a job request. ASBL depends on the list of allocated sub-meshes in the method used for selecting an allocation submesh. Results from extensive simulations under a variety of system loads confirm that the ASBL strategy improves system performance in terms of parameters such as the average turnaround times and system utilization.