Multi-objective 3D topology optimization of next generation wireless data center network

As one of the next generation network technologies for data centers, wireless data center network has important research significance. Smart architecture optimization and management are very important for wireless data center network. With the ever-increasing demand of data center resources, there are more and more data servers deployed. However, traditional wired links among servers are expensive and inflexible. Benefited from the development of intelligent optimization and other techniques, high speed wireless topology for wireless data center network is studied. Through image processing, a radio propagation model is constructed based on a heat map. The line-of-sight issue and the interference problem are also discussed. By simultaneously considering objectives of coverage, propagation intensity and interference intensity as well as the constraint of connectivity, we formulate the topology optimization problem as a multi-objective optimization problem. To seek for solutions, we employ several state-of-the-art serial MOEAs as well as three parallel MOEAs. For the grouping in distributed parallel algorithms, prior knowledge is referred. Finally, experimental results demonstrate that, the parallel MOEAs perform effectively in optimization results and efficiently in time consumption.