On fair traffic allocation and efficient utilization of network resources based on MARL

The problem of traffic balancing in a data network environment is considered in relation to Network Powered by Computing (NPC) - a new generation of computational infrastructure, where computational and data transmission resources form a network. The key contribution of this work can be formulated a...

Full description

Saved in:
Bibliographic Details
Published in:Computer networks (Amsterdam, Netherlands : 1999) Netherlands : 1999), 2024-08, Vol.250, p.110540, Article 110540
Main Authors: Stepanov, E.P., Smeliansky, R.L., Plakunov, A.V., Borisov, A.V., Zhu, Xia, Pei, Jianing, Yao, Zhen
Format: Article
Language:English
Subjects:
MARL
NPC
Traffic engineering
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The problem of traffic balancing in a data network environment is considered in relation to Network Powered by Computing (NPC) - a new generation of computational infrastructure, where computational and data transmission resources form a network. The key contribution of this work can be formulated as follows: the combination of multi-agent optimization technique with reinforcement learning and consistent hashing method ensures efficient use of bandwidth in the overlay network. This combination, named Multi-Agent Routing with Hashing (MAROH), demonstrates high performance and provides fast and fair bandwidth distribution inside an overlay network. The paper provides a detailed description of MAROH and experimental results that show that MAROH exhibits better consistent flow routing balancing than ECMP and is only slightly affected by changes in network topology.
ISSN:1389-1286
1872-7069
DOI:10.1016/j.comnet.2024.110540