The MIT Alewife Machine: Architecture and Performance

Alewife is a multiprocessor architecture that supports up to 512 processing nodes connected over a scalable and costeffective mesh network at a constant cost per node. The MIT Alewife Machine, a prototype implementation of the architecture, demonstrates that a parallel system can be both scalable and programmable. Four classes of mechanisms in the machine combine to achieve these goals: software-extended coherent shared memory provides a global, linear address space; integrated message passing allows compiler and operating system designers to provide efficient communication and synchronization; latency tolerance mechanisms -- including block multithreading and prefetching -- mask unavoidable delays due to internode communication; and support for fine-grain computation allows many processors to cooperate on relatively small problem sizes. This paper shows how these mechanisms contribute to Alewife's programming model, and describes their implementation cost and performance on a 32-node ..