A binary partition-based matching algorithm for Data Distribution Management in a High-level Architecture-based distributed simulation

Abstract

Data Distribution Management (DDM) is one of the High-level Architecture (HLA) services that reduce message traffic over the network. The major purpose of DDM is to filter and route the exchange of data between federates during a federation. However, this traffic reduction usually results in a significant computational overhead, which is caused by calculating the intersection between update regions and subscription regions in a matching process. To reduce the computational overhead for the matching process, this paper proposes a binary partition-based matching algorithm for DDM in a HLA-based distributed simulation. The new matching algorithm is fundamentally based on a divide-and-conquer approach. The proposed algorithm recursively performs binary partitioning that divides the regions into two partitions that entirely cover those regions. This approach promises low computational overhead, since it does not require unnecessary comparisons within regions in different partitions. The experimental results show that the proposed algorithm performs the existing DDM-matching algorithms better and improves the scalability of the DDM.