Performance Evaluation of a Light-Tree Point-to-Multipoint Metro-Core Architecture in Correlated Traffic Scenarios

We investigate the performance of a light-tree based point-to-multipoint (P2MP) architecture that places P2MP hub transceivers in the metro-core network, under cor- related traffic conditions. The evaluation was performed using a novel algorithm that leverages the traffic correlations among the P2MP spokes/access aggregators to optimize the placement of hub transceivers and minimize the respective costs. Simula- tion results demonstrate that the algorithm performs effectively in several correlated traffic scenarios and also provides the high- est cost benefit for randomly correlated spoke traffic.