Our project is concerned with the automatic parallelization of Mercury programs. Mercury is a purely-declarative logic programming language, this makes it easy to determine whether a set of computations may be performed in parallel with one-anther. However, the problem of how to determine which computations should be executed in parallel in order to make the program perform optimally is unsolved. Therefore, our work concentrates on building a profiler-feedback automatic parallelization system for Mercury that creates programs with very good parallel performance with as little help from the programmer as possible.
