Hash functions for priority queues

Abstract

The complexity of priority queue operations is analyzed with respect to the cell probe computational model of A. Yao (J. Assoc. Comput. Mach. 28, No. 3 (1981), 615–628). A method utilizing families of hash functions is developed which permits priority queue operations to be implemented in constant worst-case time provided that a size constraint is satisfied. The minimum necessary size of a family of hash functions for computing the rank function is estimated and contrasted with the minimum size required for perfect hashing.