METHOD AND SYSTEM FOR LOAD SCHEDULING MULTIDIMENSIONAL TASKS AMONG MULTIPLE PROCESSORS

In accordance with the principles of the invention, a method and system of multiprocessor scheduling for load sharing among multiple multiprocessors to within any given error criteria, &egr;>0, is disclosed. The method comprises guessing a value for an optimal load on the processors, partitioning a plurality of multi-dimensional tasks into a set of large tasks and a set of small tasks, guessing the invention of the set of small tasks and among the set of larger tasks processors which are represented as multi-dimensional bins, packing a discretized the set of large tasks, in the guessed invention space of the multi-dimensional bins determining if the small tasks can be allocated into unallocated space of the multi-dimensional bins after packing of the discretized large tasks, and iteratively repeating the guessing steps until all the tasks can be allocated within the gussed load. The method establishes a schedule of the multi-dimensional tasks which provides a maximum load among the multi-processors which is less than (1+&egr;) times the optimal load.