Despite years of research on transport protocols, the tussle between in-network and end-to-end congestion control has not been solved. This debate is due to the variance of conditions and assumptions in different network scenarios, e.g., cellular versus data center networks. Recently, the community has proposed a few transport protocols driven by machine learning, nonetheless limited to end-to-end approaches. In this paper, we present Owl, a transport protocol based on reinforcement learning, whose goal is to select the proper congestion window learning from end-to-end features and network signals, when available. We show that our solution converges to a fair resource allocation after the learning overhead. Our kernel implementation, deployed over emulated and large scale virtual network testbeds, outperforms all benchmark solutions based on end-to-end or in-network congestion control.

Partially Oblivious Congestion Control for the Internet via Reinforcement Learning / Sacco, Alessio; Flocco, Matteo; Esposito, Flavio; Marchetto, Guido. - In: IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT. - ISSN 1932-4537. - ELETTRONICO. - (In corso di stampa), pp. 1-1. [10.1109/TNSM.2022.3215669]

Partially Oblivious Congestion Control for the Internet via Reinforcement Learning

Alessio Sacco;Guido Marchetto
In corso di stampa

Abstract

Despite years of research on transport protocols, the tussle between in-network and end-to-end congestion control has not been solved. This debate is due to the variance of conditions and assumptions in different network scenarios, e.g., cellular versus data center networks. Recently, the community has proposed a few transport protocols driven by machine learning, nonetheless limited to end-to-end approaches. In this paper, we present Owl, a transport protocol based on reinforcement learning, whose goal is to select the proper congestion window learning from end-to-end features and network signals, when available. We show that our solution converges to a fair resource allocation after the learning overhead. Our kernel implementation, deployed over emulated and large scale virtual network testbeds, outperforms all benchmark solutions based on end-to-end or in-network congestion control.
File in questo prodotto:
File Dimensione Formato  
FINAL VERSION.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 1.6 MB
Formato Adobe PDF
1.6 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2972499