All News

Check out what we have been up to in 2023.
Jan 17 2024, Laurent Vanbever

2023 was a relatively balanced year for us. Check out our activity report to get a glimpse at what we have been up to and what is in the pipeline in 2024.

Activity Report

Welcome, Laurin!
Oct 30 2023, Laurent Vanbever

Our group welcomes Laurin as new team member. Before starting as a PhD student, Laurin did his master in computer sciences at ETH and worked on serverless architectures and associative caches, amongst others. Welcome!

Welcome, Laurin!
Machine Learning for Networking: What? With what? For what?
Oct 30 2023, Laurent Vanbever

A couple of weeks ago I gave a talk at the Google Networking Summit on some possible applications of machine learning to networking problems. The talk looked in turn at: (i) what kind of models we should learn (hint: transformers-based models); (ii) how we can get our hands on network data to train these models (hint: leveraging big code!); and (iii) how much networking knowledge do large-language models have nowadays (hint: they’re pretty good, actually). You can find the slides here.

Slides

QVISOR accepted at HotNets '23!
Sep 8 2023, Laurent Vanbever

Our paper “QVISOR: Virtualizing Packet Scheduling Policies” has been accepted at ACM HotNets 2023! In this work, we ask ourselves: is it possible to simultaneously deploy multiple scheduling algorithms on existing commodity switches? Take a look at the pre-print to find out!

We have a new website!
Jun 2 2023, Tibor Schneider

We finally have a new group website with a new design. There is now a full-text search for the publication list (which has grown so large that it is difficult to find a specific publication). And, of course, there is a dark-mode!

Upcoming SIGCOMM '23 paper!
May 19 2023, Laurent Vanbever

Happy to report that our group will again be represented at SIGCOMM this year! Our paper on seamless network configurations, the first avoiding both permanent and transient violations, has just been accepted. As usual, stay tuned for the details! New York City, here we come :-)

Upcoming SIGCOMM '23 paper!
Welcome, Theo!
May 1 2023, Laurent Vanbever

Our group welcomes Theo as new team member. Before starting as a PhD student, Theo did his master thesis on routing attacks in our group which resulted in a publication at EuroS&P. Check it out!

Welcome, Theo!
Check out what we have been up to in 2022.
Jan 10 2023, Laurent Vanbever

2022 was one of our best year ever, on many accounts. Check out our activity report to see what our group has been up to and what is in the tank for us for 2023.

Activity Report

New paper accepted at EuroP4 2022!
Oct 14 2022, Laurent Vanbever

Our paper “Reducing P4 Language’s Voluminosity using Higher-Level Constructs” has been accepted at EuroP4 2022! In this paper, we present O4, an extension of P4, that incorporates three higher-level constructs (arrays, loops, and factories) to reduce the voluminosity of P4 code.

New paper accepted at EuroP4 2022!
New paper accepted at NeurIPS 2022!
Sep 15 2022, Laurent Vanbever

Our paper entitled “Learning to Configure Computer Networks with Neural Algorithmic Reasoning” was accepted at NeurIPS 2022! In this paper, we explain how we can approximate routing computations using neural networks. Among others, doing so allows us to efficiently “invert” these computations enabling to automatically synthesize configurations from their intended output. This synthesis problem is known to be hard: actually, our recent ICNP 2022 paper shows that many instances of that problem are NP-hard/NP-complete. Having a away to approximate these computations allows us to “break” the inherent scalability barrier of solving these problems, at the price of accuracy. How to deal with this accuracy loss is amongst the many next questions we want to look at. Stay tuned!

New paper accepted at NeurIPS 2022!
Two papers accepted at HotNets 2022!
Sep 5 2022, Laurent Vanbever

Our group will have two papers at this upcoming ACM HotNets workshop! These two papers will mark our 10th and 11th HotNets papers since 2014.

Stay tuned to learn more about:

  1. How we plan to build the next-generation of network traffic generator by leveraging millions of code repositories hosted on code-sharing platforms such as GitHub;

and

  1. How we intend to build generalizable machine learning (ML) models for predicting network traffic dynamics using the Transformer architecture.

As usual, you’ll find the final version of the papers on our publications page in a couple of weeks.