Optimal rate and aggregate bandwidth allocation under service performance requirements in data center backbone networks

Recent years have seen an increasing demand for multiclass services with different performance requirements in data center backbone networks (DCBNs). Unlike the simple multi-path routing schemes to balance the load, an enhanced network utility maximization (NUM) framework is proposed to optimize rate and aggregate bandwidth allocation for DCBN in the paper, with constraints on bandwidth resource and multiple performance requirements which include packet delay, loss, and traffic throughput. Because of the non-convexity of some constraint functions, the formulated NUM problem is non-convex, which is challenging to solve. To deal with such a problem, we propose sequential parametric convex approximation (SPCA) to transform it into several convex sub-problems. Then we obtain optimal solution of the formulated problem by solving the convex ones above, using a well-known available tool. We validate our method via simulations on the topology of Google's datacenter backbone network. Numerical results show that the proposed method guarantees that the optimal rate and aggregate bandwidth allocation for each traffic class strictly meets its every performance requirement, yet runs 1.5 times faster than the state-of-art MCTEQ algorithm.