A message passing implementation of lazy task creation

Abstract

This paper describes an implementation technique for Multilisp's future construct aimed at large shared-memory multiprocessors. The technique is a variant of lazy task creation. The original implementation of lazy task creation described in [Mohr, 199l] relies on efficient shared memory to distribute tasks between processors. In contrast, we propose a task distribution method based on a message passing paradigm. Its main advantages are that it is simpler to implement, has a lower cost for locally run tasks, and allows full caching of the stack on cache incoherent machines. Benchmarks on a 32 processor BBN TC2000 show that our method is more efficient than the original implementation by as much as a factor of 2.