Studies the execution of a Fortran DO loop, modeled as aniteration space and several dependence vectors, on an asynchronousshared-memory multiprocessor, where each iteration is the basic unit tobe executed by a single processing element. During the execution of a DOloop, an iteration will pass through four states, namely, idle, pending,ready, and finished states. By describing an execution scheme, the paperstudies (1) the number of iterations not depending on any otheriterations, (2) the maximum number of ready iterations at any instanceduring the execution, (3) the maximum number of pending iterations atany instance during the execution, (4) a hash function to dispersedifferent pending iterations, and (5) the parallel execution time
展开▼