Dynamic erasure-coded data retrieval in elastic optical data center networks

Abstract

A flexible data retrieval service is needed in data center networks, where erasure coding has been widely adopted. Elastic Optical Networks, based on Nyquist wavelength division multiplexing or optical orthogonal frequency division multiplexing is an ideal backbone for data centers, since they can efficiently utilize the optical fiber's bandwidth in an elastic manner by partitioning the bandwidth into hundreds or even thousands of subcarriers. In this paper, we study the erasure-coded data retrieval problem for dynamic traffic in elastic optical networks, the objective is to minimize the blocking ratio. A heuristic based on List Scheduling was proposed. Simulation results show that the blocking ratio is reduced in comparison to a benchmark method.