Iris is a system to coordinate complex network measurements from multiple vantage points. Think of it as a project similar to CAIDA Ark or RIPE Atlas, with the following features:
- Fully open-source code.
- Handle multi-round measurements, such as diamond-miner IP tracing measurements.
- Handle both centralized computation on a powerful server, and distributed probing on smaller agents.
- Can tolerate the temporary loss of agents, database or control-plane components.
Iris generates IP Route Survey (IPRS) snapshots, which are published through Measurement Lab (M-Lab) and are publicly accessible.
@article{10.1145/3523230.3523232,
author = {Gouel, Matthieu and Vermeulen, Kevin and Mouchet, Maxime and Rohrer, Justin P. and Fourmaux, Olivier and Friedman, Timur},
title = {Zeph & Iris Map the Internet: A Resilient Reinforcement Learning Approach to Distributed IP Route Tracing},
year = {2022},
issue_date = {January 2022},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
volume = {52},
number = {1},
issn = {0146-4833},
url = {https://doi.org/10.1145/3523230.3523232},
doi = {10.1145/3523230.3523232},
journal = {SIGCOMM Comput. Commun. Rev.},
month = {mar},
pages = {2–9},
numpages = {8},
keywords = {active internet measurements, internet topology}
}Iris is developed and maintained by the Dioptra group at Sorbonne Université in Paris, France.