Static multi-sourced data retrieval in elastic optical networks
Journal of Optical Communications and Networking
Extreme-scale science applications are highly innovative and constantly evolving. They are expected to generate data in the petabyte and exabyte ranges. Erasure coding has been widely adopted for data storage in data center networks, where the data are encoded and stored in multiple locations. Therefore, an efficient data retrieval service is needed to transfer encoded data from selected multiple stored nodes to a single destination. Elastic optical networks are a promising backbone technology for data center communication due to their capability to efficiently and flexibly allocate the huge optical bandwidth to heterogeneous traffic demands. In this paper, the erasure-coded multi-sourced data retrieval routing and scheduling problem is studied for static traffic in elastic optical networks, and the objective is to minimize the total transmission completion time of all the requests. An integer linear programming formulation and low-complexity heuristic are proposed. Furthermore, analytical lower bounds are derived and a meta-heuristic, Tabu Search, is adopted to solve the problem. Numerical results are presented to show the effectiveness of the proposed methods.
National Science Foundation
Juzi Zhao and Vinod M. Vokkarane. "Static multi-sourced data retrieval in elastic optical networks" Journal of Optical Communications and Networking (2022): 792-804. https://doi.org/10.1364/JOCN.465019