A Comparative Study of Data Distribution Management Algorithms

Abstract

In large-scale distributed defense modeling and simulation, Data Distribution Management (DDM) controls and limits the data exchanged reducing the processing requirements of federates. In this paper, we present a comparative study of a recently proposed DDM algorithm, called P-Pruning algorithm, with three other known techniques: region-matching, fixed-grid, and dynamic-grid DDM algorithms. By populating the multicast group, first only on the basis of X-axis information of routing space, and pruning the non-overlapping subscriber regions within multicast groups in successive steps, the P-Pruning algorithm avoids the computational overheads of other algorithms. From the simulation study, we found that the P-Pruning algorithm is faster than the other three DDM algorithms. The performance evaluation results also show that the P-Pruning DDM algorithm uses memory at run-time more efficiently and requires less number of multicast groups as compared to the three algorithms. We also present the design and implementation of a memory-efficient, scalable enhancement to the P-Pruning algorithm.