,
Ayush Mishra
,
Laurent Vanbever
Creative Commons Attribution 4.0 International license
Our current view of traffic on the Internet - the world’s largest and most pervasive network - comes from a variety of perspectives, each with its own blind spots and biases. In this paper, we make the case for using publicly available Internet exchange point (IXP) statistics as a complementary vantage point. While IXP data has its own limitations, it is fine-grained, accessible, and independently verifiable - offering a distinct perspective on Internet usage patterns. We present results from a two-year study (2023-2024) of 472 IXPs worldwide, capturing approximately 300 Tbps of peak daily aggregate traffic by late 2024. Over this period, aggregate IXP traffic increased by 49.2% (24.5% annualized), with regionally distinct diurnal patterns and event-driven anomalies. These results provide an accessible framework for researchers and operators to study the Internet’s evolving ecosystem from an IXP-based perspective, and lay the groundwork for systematic, global-scale detection of network anomalies and outages.
@InProceedings{kirci_et_al:OASIcs.NINeS.2026.25,
author = {Kirci, Ege Cem and Mishra, Ayush and Vanbever, Laurent},
title = {{Five Blind Men and the Internet: Towards an Understanding of Internet Traffic}},
booktitle = {1st New Ideas in Networked Systems (NINeS 2026)},
pages = {25:1--25:26},
series = {Open Access Series in Informatics (OASIcs)},
ISBN = {978-3-95977-414-7},
ISSN = {2190-6807},
year = {2026},
volume = {139},
editor = {Argyraki, Katerina and Panda, Aurojit},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.NINeS.2026.25},
URN = {urn:nbn:de:0030-drops-256106},
doi = {10.4230/OASIcs.NINeS.2026.25},
annote = {Keywords: Internet Exchange Point (IXP), traffic measurement, longitudinal study, traffic growth, diurnal patterns, PeeringDB, global-scale detection, network anomalies}
}