Title: Inferring business relationships in the internet backbone

Authors: Martin Huth; Benjamin Fabian

Addresses: Institute of Information Systems, Humboldt University Berlin, Spandauer Str. 1, 10178 Berlin, Germany ' Institute of Information Systems, Humboldt University Berlin, Spandauer Str. 1, 10178 Berlin, Germany

Abstract: Several researchers during the last decade have encountered the problem of how to infer business relationships between autonomous systems (ASes) of the internet. Since the internet has a decentralised structure and public data sources containing inter-domain routing information have not been created for topology inference, there are no accurate and comprehensive maps of the internet readily accessible. This challenge has inspired several approaches for inferring business relationships between ASes from BGP routing data. This article presents one implementation of the most recent and most promising approach for relationship inference on AS-level. The algorithm used has been improved in terms performance and quality of the sanitising process. Unlike recent projects, not a only snapshot of the topology of the internet has been inferred but a comprehensive map showing the internet over the last decade. The correctness of this implementation and the inferred dataset is examined by comparison with a business relationship graph and a validation dataset provided by related work.

Keywords: internet topology; business relationships; autonomous systems; internet measurement; topology inference; internet maps.

DOI: 10.1504/IJNVO.2016.081651

International Journal of Networking and Virtual Organisations, 2016 Vol.16 No.4, pp.315 - 345

Received: 04 Nov 2015
Accepted: 08 Dec 2015
Published online: 19 Jan 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article