Design and implementation of efficient control for incoming inter-domain traffic with Information-Centric Networking. (1st May 2019)

• Record Type: Journal Article
• Authors: Li, Jiawei; Luo, Hongbin; Zhang, Shan; Li, Hongyi; Yan, Fei
• Is Part Of: Journal of network and computer applications. Volume 133(2019)
• Page Start: 109
• Page End: 125
• Publication Date: 2019-05-01
• Subjects: Information-Centric Networking(ICN) -- Traffic Engineering; Microcomputers -- Periodicals; Computer networks -- Periodicals; Application software -- Periodicals; Micro-ordinateurs -- Périodiques; Réseaux d'ordinateurs -- Périodiques; Logiciels d'application -- Périodiques; Application software; Computer networks; Microcomputers; Periodicals; 004.05; 004
• DOI: 10.1016/j.jnca.2019.02.013
• Languages: English
• ISSNs: 1084-8045

Abstract

Abstract: Efficient and accurate control of incoming inter-domain traffic is critically important for Autonomous Systems (ASes). However, it is extremely difficult for two neighboring ASes to efficiently control incoming traffic over multiple inter-domain links in the current Internet. By contrast, Information-Centric Networking (ICN) is a novel network paradigm and has many inherent features, which can support finer granularity of routing policies and provide more opportunities to realize Traffic Engineering (TE). In this paper, we explore the potentials of ICN in controlling incoming inter-domain traffic between neighboring ASes who have multiple inter-domain links. We present a heuristic algorithm and extend it with different information on network status and content metadata. We implement them in a real prototype and compare them with a classic inter-domain TE approach in the current Internet (i.e., the IP-prefixes negotiation approach, IPN) in three scenarios. Three interesting conclusions are drawn from extensive experiments on the prototype. First, ICN makes it easier to control inter-domain traffic compared with IPN. In particular, ICN approaches have shorter negotiation delay and lower negotiation overhead. Second, content metadata (e.g., content size) in ICN is quite beneficial to control traffic since it can help estimate traffic volumes so as to achieve more accurate control of traffic. Third, ICN can efficiently handle the bursty traffic and link failures compared with IPN.