Bandwidth scheduling for big data transfer using multiple fixed node-disjoint paths

Many large-scale applications require the transfer of big data over high-performance networks for remote operations. Such requirements call for a fast bandwidth scheduling solution to discover feasible and efficient reservation options in network environments with time-varying bandwidths. We formulate a generic problem of Bandwidth Scheduling with Two Node-Disjoint Paths (BS-2NDP) to support big data transfer. In BS-2NDP, we further consider two different types of paths: (i) two fixed paths with fixed bandwidth (2FPFB), and (ii) two fixed paths with variable bandwidth (2FPVB). We prove that both 2FPFB and 2FPVB are NP-complete, and design a heuristic approach for each of them. We implement and evaluate these scheduling algorithms in both simulated and real-life networks. Extensive results show that the proposed heuristics achieve a close-to-optimal performance in small-scale networks, and significantly outperform other heuristic approaches in large-scale networks.