Adaptive Parallelism on Multiprocessors: Preliminary Experience with Piranha on the CM-5

Mechanisms for sharing multiprocessors among users are still in their infancy|typical approaches include simple space-sharing and ineecient, restricted forms of time-sharing. In this work we investigate a new alternative: adaptive parallelism 2]. Adaptively parallel programs can execute over a dynamically changing set of processors; many such codes can easily and dynamically share a multiprocessor by individually adapting to execute in separate groups of processors that may vary with time. Adaptive parallelism has been successfully used in Pi-ranha, an execution model for Linda 1 programs that turns idle networked workstations into a signiicant computing resource 4]. This work explores Piranha on the Connection Machine CM-5 multiprocessor. Our preliminary results suggest that adaptive parallelism provides not only substantial computing power for parallel programs but also an attractive alternative to traditional methods for sharing multiprocessors among users.