Automatic parallelization of sequential code for a cluster of multiprocessors

In recent years parallel architectures have rapidly evolved to solve increasingly large problems, mainly because single processor based systems have reached the limit of their physical capacity. The end of 90’s has seen the growth and evolution of parallel architectures towards a cluster of multiprocessors owing to scaling problems with shared memory machines. The efficient use of such parallel architectures demands writing of parallel code by the programmer or automatic parallelization of sequential code by the compiler. We shall look at the latter approach in more detail since the former places too much responsibility on the individual. Our approach is based on SUIF, an automatic parallelizer for a symmetric memory architecture, developed by the Stanford University. We shall focus on extending SUIF to produce high level code with explicit calls to parallel libraries like MPI instead of calls to runtime libraries. This would allow us to produce parallel code for a cluster of multiprocessors instead of a shared memory machine.