Managing distributed storage system through network redesign

In this paper, we have assessed the performance of an existing enterprise network before and after deploying distributed storage systems through a co-simulation environment integrating NS-2 with Simulink. The existing enterprise networks can benefit from replacing the expensive monolithic disk array systems with low-cost and scalable distributed storage systems. The storage replacement task is a challenging redesign problem, where we need to ensure the suitability and performance sustainability of existing network to support distributed storage systems. We have chosen three performance parameters such as average response time, packet loss rate and storage system throughput as the outcomes of the co-simulation, and these parameters can guide the redesign process by maximizing the storage system throughput with the intuition to reduce average response time and packet loss rate. The simulation of an enterprise network with 680 clients and 54 servers followed by redesigning shows improvements in the storage system throughput by 13.9% with 24.4% decreased average response time and 38.3% reduced packet loss rate.