The shared-thread multiprocessor

This paper describes initial results for an architecture called the Shared-Thread Multiprocessor (STMP). The STMP com-bines features of a multithreaded processor and a chip mul-tiprocessor; specically, it enables distinct cores on a chip multiprocessor to share thread state. This shared thread state allows the system to schedule threads from a shared pool onto individual cores, allowing for rapid movement of threads between cores. This paper demonstrates and evaluates three benets of this architecture: (1) By providing more thread state stor-age than available in the cores themselves, the architecture enjoys the ILP benets of many threads, but carries the in-core complexity of supporting just a few. (2) Threads can move between cores fast enough to hide long-latency events such as memory accesses. This enables very-short-term load balancing in response to such events. (3) The system can redistribute threads to maximize symbiotic be-havior and balance load much more often than traditional operating system thread scheduling and context switching