Mapping applications to NoC platforms with multithreaded processor resources

Network on chip (NoC) is a new design paradigm for building scalable core-based systems on chip (SoC). Multithreading is a technique for hiding long latencies of memory accesses, through the overlapped execution of several threads. In this paper, we make a case for using multi-threaded processors (MTP) as resources in NoC and describe a methodology for off-line mapping and scheduling of concurrent applications to NoC with MTPs. The experimental results show a speedup of 15% on average for a 2/spl times/2 NoC when using two MTPs with 3-thread contexts instead of two general processors (GP).