Dynamic Dependence Analysis: A Novel Method for Data Depndence Evaluation

This paper analyzes the results of applying data dependence calculations to the Perfect Benchmarks suite. We show that it is possible to measure the optimal parallelism, as deened by our model, and to compare the obtained parallelism for various data dependence tests with the optimal parallelism. Using these techniques, we nd that a variation of Banerjee's inequalities is suucient in all cases to obtain more than half of the available parallelism, and that a full integer-programming based dependence test does not contribute signiicantly to the measured parallelism.