Automated transformation of sequential divide-and-conquer algorithms into parallel programs

摘要: Divide-and-conquer algorithms obtain the solution to a given problem by dividing it into subproblems, solving these recursively and combining their solutions. In this paper we present system that automatically transforms sequential divide-and-conquer written in C programming language parallel code which is then executed on message-passing multicomputers. The user of expected add only few annotations an existing program. strategies required for transforming source executable binaries are discussed. performance speedups attainable will be illustrated several examples.