Elisabeth Brunet

Elisabeth Brunet

Associate Professor in Computer Science

Phone: +33 1 60 76 47 40
Email: elisabeth.brunet@telecom-sudparis.eu




  • High Performance Computing Systems
  • High Performance Communications
  • GPU Algorithms
  • Applied Machine Learning
  • Frugal Computing

Current Ph.D. Students

List of former students accessible here.


Recent work:

  • Comparing SyCL Data-Transfer Strategies for Tracking Use Cases. Sylvain Joube, Hadrien Grasland, David Chamont, Elisabeth Brunet. ACAT 2021, to appear in the Journal of Physics: Conference Series (JPCS)
  • Transparent Overlapping of Blocking Communication in MPI Applications. Alexis Lescouet, Elisabeth Brunet, François Trahay, Gaël Thomas. HPCC/DSS/SmartCity 2020
  • Using Differential Execution Analysis to Identify Thread Interference. Mohamed Said Mosli Bouksiaa, François Trahay, Alexis Lescouet, Gauthier Voron, Rémi Dulong, Amina Guermouche, Elisabeth Brunet, Gaël Thomas. IEEE Trans. Parallel Distributed Syst. 30(12), 2019

Complete list of publications accessible here.


Past teaching activities accessible here.


I am an associate professor at Telecom SudParis in the Parallel & Distributed Systems group since 2010. Previously, I spent one year as a postdoc in the Joint Laboratory of Petascale Computing of the University of Illinois at Urbana-Champaign. I did my PhD thesis within the INRIA Runtime team of LaBRI (University of Bordeaux, France) under the direction of Raymond Namyst and defended it in December 2008.

Initially, my primary research interests are in high performance computing in the cluster architecture landscape. More particularly, my work is organized around high performance communication, high speed network exploitation, design of dynamic optimized communication scheduling from generic algorithms and current high performance communication libraries such as MPI implementations. Then, I focused my research on compilers for a while in order to automatically transform parallel program in distributed ones with the ambition to enhance their scalability. More recently, I have been interested in a transversal project with biologists and physicians that deals with the increase of accurate data generation and their analysis in order to help biologists to answer fundamental questions. In this context, my interests are right now focused on AI runtimes optimization in order to treat pretty large histological sections on parallel frugal hardware.


Email: elisabeth.brunet@telecom-sudparis.eu

Web: ebrunet.wp.imtbs-tsp.eu

Address: Telecom SudParis, 19 place Marguerite Perey 91120 Palaiseau France, Office 4A320

Tel.: +33 1 75 31 44 32