Deep reinforcement learning for smart queue management - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Deep reinforcement learning for smart queue management

Résumé

With the goal of meeting the stringent throughput and delay requirements of classified network flows, we propose a Deep Q-learning Network (DQN) for optimal weight selection in an active queue management system based on Weighted Fair Queuing (WFQ). Our system schedules flows belonging to different priority classes (Gold, Silver, and Bronze) into separate queues, and learns how and when to dequeue from each queue. The neural network implements deep reinforcement learning tools such as target networks and replay buffers to help learn the best weights depending on the network state. We show, via simulations, that our algorithm converges to an efficient model capable of adapting to the flow demands, producing thus lower delays with respect to traditional WFQ.
Fichier non déposé

Dates et versions

hal-03546621 , version 1 (28-01-2022)

Identifiants

Citer

Hassan Fawaz, Djamal Zeghlache, Quang Tran Anh Pham, Leguay Jérémie, Medagliani Paolo. Deep reinforcement learning for smart queue management. NETSYS 2021: Conference on Networked Systems 2021, TU Berlin, Sep 2021, Lübeck (online), Germany. pp.1-14, ⟨10.14279/tuj.eceasst.80.1139⟩. ⟨hal-03546621⟩
85 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More