Lower bounds on the loading of degree-2 multiple bus networks for binary-tree algorithms

Abstract

A algorithm, Bin(n), proceeds level-by-level from the leaves of a 2/sup n/-leaf balanced binary tree to its root. This paper deals with running algorithms on multiple bus networks (MBNs) in which processors communicate via buses. Every binary-tree has a degree (maximum number of buses connected to a processor) of at least 2. There exists a degree-2 MBN for Bin(n) that has a loading (maximum number of processors connected to a bus) of /spl theta/ (n). For any MBN that runs Bin(n) optimally, the loading was recently proved to be /spl Omega/(n 1/2 ). In this paper, we narrow the gap between previous results by deriving a tighter lower bound of /spl Omega/(n2/3). We also establish a tradeoff between the speed and loading of degree-2 MBNs.