Effective Scheduling Function Design in SDN through Deep Reinforcement Learning

TL;DR

This paper introduces a deep reinforcement learning method to automatically design a general and effective scheduling function for SDN controllers, outperforming traditional heuristics across diverse network scenarios.

Contribution

It proposes a neural network-based scheduling function and a novel training approach using deep RL, enabling automatic, adaptable, and high-performance request dispatching in SDN.

Findings

RL-designed SF outperforms heuristics in simulations

The trained SF generalizes well across network settings

Rapid learning of optimal scheduling functions achieved

Abstract

Recent research on Software-Defined Networking (SDN) strongly promotes the adoption of distributed controller architectures. To achieve high network performance, designing a scheduling function (SF) to properly dispatch requests from each switch to suitable controllers becomes critical. However, existing literature tends to design the SF targeted at specific network settings. In this paper, a reinforcement-learning-based (RL) approach is proposed with the aim to automatically learn a general, effective, and efficient SF. In particular, a new dispatching system is introduced in which the SF is represented as a neural network that determines the priority of each controller. Based on the priorities, a controller is selected using our proposed probability selection scheme to balance the trade-off between exploration and exploitation during learning. In order to train a general SF, we first formulate the scheduling function design problem as an RL problem. Then a new training approach is developed based on a state-of-the-art deep RL algorithm. Our simulation results show that our RL approach can rapidly design (or learn) SFs with optimal performance. Apart from that, the trained SF can generalize well and outperforms commonly used scheduling heuristics under various network settings.