An application of program unification to priority queue vectorization

Abstract

In this experimental study, we apply the technique of program unification to priority queues. We examine the performance of a variety of unified priority queue implementations on a Cray Y-MP. The scope of the study is restricted to determining if different implementations of priority queues exhibit markedly different performance characteristics under program unification. We found this to be true. In a larger view, this result has interesting consequences in the application of program unification to discrete event simulations on vector or SIMD machines. We find the heap to be a promising data structure in the program unification paradigm.

Experiments with Program unification on the Cray Y‐MP

“…In one case, performance improved by a factor of roughly 8. In combination with previous work [31,32,34], the results of this study suggest that program unification is a viable and practical technique for enhancing program performance on vector machines. Our current work involves more detailed empirical studies of program unification and related scheduling issues on vector/SIMD machines, and a program unification tool for both architectures.…”

“…We use the heap[33] to implement a priority queue because it is known to be an efficient structure for simulation event processing. More importantly, we have found it to exhibit excellent performance when unified and executed in a general setting [34]. A sequential program for the single-server queueing system simulation begins with system initialization and generation of the first customer arrival.…”