Modeling and simulation of network redesign through clusters consolidation

We have proposed a design and analysis tool for network topology, where it combined a custom-made design/redesign methodology and a commercial simulation program, OPNET modeler. Here we assume an existing network with C clusters as a reconfigurable architecture and we have proposed a number of redesign techniques to reduce the extra-traffic and maximize the intra-traffic, the traffic within the clusters (subnetworks). We have utilized genetic algorithms (GA) to search the network size of N customers, distributed in an area of 100 m times100 m to optimize each redesign operation. We analyzed the performance of the network before and after implementing the redesign at the customer-level using OPNET modeler. Our simulation study using OPNET shows a better trade-off by increasing the utilization of the redesigned network around 19%, thereby reducing the network cost. This helps for capacity planning of an existing network prior to the deployment of redesign operations in a real-time network.