Grid-based data management in distributed simulation

Abstract

The data distribution management (DDM) service provided by the High Level Architecture Run-time Infrastructure is aimed at making data communication more efficient by sending data only to those federates who need the data, as opposed to the broadcasting mechanism employed by Distributed Interactive Simulation. Several DDM methods, including class, region and grid-based, have appeared in the literature. This paper discusses grid-based DDM and presents a simulation platform developed and implemented for investigating the impact of grid-based filtering on performance. Our experimental results show that the choice of the grid-cell size has a substantial impact on the performance of distributed simulation. The optimal cell size depends on the characteristics of the application and the underlying computer system. In many practical simulations, it is not easy to determine the optimal cell size. Our experience indicates that in such a case one should use larger cells rather than very small ones.