ACM SIGCOMM 2022 TUTORIAL: Large-scale Open Internet Measurement with M-Lab
Monday August 22, 2022 9:00 am - 12:30 pm CEST (Room: Mendes da Costa kamer)
9:00 am - 9:30 am CEST Welcome + Intro to M-Lab
- Welcome + Intro to M-Lab
9:00 am - 9:30 am CEST
Introduction to M-Lab
Presenter: Lai Yi Ohlsen (M-Lab) / Phillipa Gill (Google)
This introduction will cover the history of M-Lab and what it offers the network measurement community. It will cover what the platform offers for new network measurements as well as give an overview of data sets available from M-Lab.
9:30 am - 10:00 am CEST M-Lab in Practice
- M-Lab in Practice
9:30 am - 9:45 am CEST
Presenter: Kévin Vermeulen (CNRS)
9:45 am - 10:00 am CEST
Network Tomography with M-Lab Traceroute Data
Presenter: Zeinab Shmeis (EPFL)
10:00 am - 10:30 am CEST Introduction to Hands on Activity
- Introduction to Hands on Activity
10:00 am - 10:30 am CEST
Introduction to Hands on Activity
Presenter: Roberto D'Auria (M-Lab)
10:30 am - 11:00 am CEST Break
11:00 am - 11:30 am CEST Hands on Activity
11:00 am - 11:30 am CEST
We will provide the attendees with a Colab notebook with an example query/starter code. We will walk the attendees through running queries on different M-Lab datasets using Colab.
11:30 am - 12:30 pm CEST Panel Discussion
11:30 am - 12:30 pm CEST
Current challenges in measuring Internet connectivity
Panelists: Timur Friedman (Sorbonne U.), Amreesh Phokeer (ISOC), Renata Teixeira (Netflix), Kévin Vermeulen (CNRS)
After laying the foundation with the introduction and hands on activity we will have a panel discussion with researchers working on measuring Internet connectivity to discuss challenges in this space.
Call For Participation
Measurement Lab (M-Lab) is an open, distributed server platform on which researchers can deploy different network
measurement services. The project is dedicated to: (1) Providing an open, verifiable measurement platform for
global network measurement, (2) Hosting the largest open Internet performance dataset on the planet and (3)
Creating visualizations and tools to help people make sense of Internet performance. All the data collected by the
M-Lab platform is openly available and all measurement tools hosted on the platform are open source.
Since its inception over a decade ago, the platform has grown and developed in ways that make its data and resources easier to use. Data is now stored in BigQuery tables that allow easy querying and analysis of the data from cloud-based Colab notebooks. 2019 also saw the transition of M-Lab from using a PlanetLab-compatible resource “slicing” approach, to a more modern approach for deploying and managing experiments. The M-Lab servers are administered and maintained as a managed Kubernetes cluster. Experiments are deployed to servers using Docker containers.
Goal. Through this tutorial we plan to make the networking community more aware of our data and tools that make leveraging M-Lab for research and education easier than ever before. Through the tutorial we hope to achieve the following goals:
- Raise awareness of M-Lab data. Through a hands-on session we hope to increase awareness of the different types of data available from M-Lab. Participants will learn how M-Lab data can be used in both research and classroom settings.
- Build community. Through invited talks from researchers that have used M-Lab data we aim to build community among researchers using our data to share tools and techniques that have enabled successful research with the platform.
- Expose open research challenges. Via a panel discussion, we plan to discuss research challenges associated with measuring the quality of users’ Internet connectivity. We hope this discussion will encourage new lines of research about how to measure network connectivity at scale.
Audience Expectations and Prerequisites
Attendees will need a modern Web browser to run Colab. They will also need to sign up for access to the M-Lab BigQuery data (by joining firstname.lastname@example.org mailing list). We will send specific technical specifications for browsers to run Colab and instructions to gain access to the data ahead of the tutorial.
Phillipa Gill is a Research Scientist at Google where she leads the Internet Research Team as they contribute to the Measurement Lab project. Prior to joining Google she was an Associate Professor at the University of Massachusetts--Amherst. She holds a PhD from the University of Toronto. She was named one of the Top Innovators Under 35 by MIT Tech Review, and one of the 10 Women in Networking/Communications That You Should Watch by N2Women.
Lai Yi Ohlsen
Lai Yi Ohlsen is a tech worker and artist, and the Director of Measurement Lab, one of Code for Science Society's sponsored projects. Previously, she worked to defend and promote human rights online with eQualitie as Technical Programs Manager. Ohlsen is also a 2019 Movement Research Artist in Residence and has been supported by the Pioneer Works Technology Lab and the Internet Archive’s Decentralized Web Summit. With a background in computer science, project management and community organizing she translates technical visions into actionable, accessible, and assessable plans. Her multidisciplinary work across industries aims to better understand who and what technology is for.