ECO: Efficient Collective Operations for communication on heterogeneous networks

PVM and other distributed computing systems have enabled the use of networks of workstations for parallel computation, but their approach of treating all networks as collections of point-to-point connections does not promote efficient communication-particularly collective communication. The Efficient Collective Operations (ECO) package contains programs which solve this problem by analyzing the network and establishing efficient communication patterns. This paper describes ECO and gives performance results of using ECO to implement the collective communication in CHARMM, a widely used macromolecular dynamics package. ECO substantially improves the performance of CHARMM on a heterogeneous network. ECO's approach gives a programmer the ability to use the available networks to their full potential without acquiring any knowledge of the network structure.