Parallel processing of near fine grain tasks using static scheduling on OSCAR (optimally scheduled advanced multiprocessor)
Hironori Kasahara, Hiroki Honda, Seinosuke Narita · Conference on High Performance Computing (Supercomputing) · 1990
This paper proposes a compilation scheme for parallel processing of near fine grain tasks, each of which consists of several instructions or a statement, on a multiprocessor system called OSCAR (Optimally Scheduled Advanced Multiprocessor). The scheme allows us to minimize synchronization and data transfer overheads and to optimally use registers of each processor by using a static scheduling algorithm considering data transfer. This scheme can effectively be combined with macro-dataflow computation and the loop concurrentization. A compiler using the proposed scheme has been implemented on OSCAR, which has been designed to take full advantage of the static scheduling. In this paper, performance evaluation of the scheme on OSCAR is also described.