Introducing PSO for Optimal Packet Scheduling of Collective Communication

Abstract

Interconnection network is an inevitable component that is responsible to the system's communication capability. It affects the system-level performance as well as the physical and logical structure of the parallel system. Many studies are reported to enhance the interconnection network technology, however, we have to further discuss remaining issues for building large-scale systems. One of the most important issues is congestion management. In an interconnection network, packets are transferred simultaneously, and the packets interfere to each other on the network. Congestion arises as a result of the interference among packets. Its fast spreading speed degrades communication performance drastically and it continues for long time. Thus, we should appropriately control the network to suppress the congested situation for maintaining the maximum performance. Many studies address the problem and present effective methods, however, the maximal performance in an ideal situation is not sufficiently clarified. Solving the ideal performance is, in general, an NP-hard problem. This paper introduces particle swarm optimization (PSO) method to overcome the problem. In this paper, we first formalize the optimization problem suitable for the PSO method and present three PSO methods for avoiding local minima. We furthermore introduce some non-PSO methods for comparison. Our preliminary evaluation results reveal high potentials of the PSO method.