Network state information aided scheduling algorithms with statistical guarantees

Statistical guarantee can dramatically increase the utilization of link bandwidth compared with deterministic guarantee. With aiding of network state information, scheduling schemes can reduce the probability of end-to-end delay violation and improve the overall performance. In this paper, we first investigate the single node performance of EDF (earliest deadline first) scheduler with statistical guarantees. Then based on these results, we propose two network state information aided scheduling mechanisms, named as DP-EDF and DDP-EDF, and validate their advantage over traditional EDF by both theoretical analysis and simulations.