Transformation method to reduce loop overhead in HPF compiler

Among data parallel languages, which were suggested to solve programming problems of distributed memory machines, HPF is regarded as a standard data parallel language. And most HPF compilers are source-to-source translators because they can be implemented easily. By the way, they currently generate significant amount of ineffective codes. In particular, FORALL construct is converted into several DO loops, so loop overhead, which exacerbates program performance, of these codes is very increased. Therefore HPF source-to-source compilers require some techniques to produce optimized DO loop from FORALL construct. In this paper, we propose transformation method to reduce loop overhead for FORALL construct. For this, we define and use relation distance vector which can represent both data dependence information and flow information. Then we evaluate and analyze execution time for the codes converted by our method and by PARADIGM method in aspect of array size and dependence distance.