Projects/events

Current projects/events

• I am member of the NumPEX project on ExaScale. I am member of the integrated project ExaDOST on data management and I am responsible for the international collaborations.
• I am member of the Joint Laboratory for Extreme-Scale Computing between INRIA and the University of Illinois at Urbana Champaign, Argonne National Lab, BSC, Univ of Tennessee, Riken, Julich and Inria. The joint lab focuses on software challenges found in complex high-performance computers.
• I am member of the steering committee of the ICPP Conference since 2022.

Past projects

• The main objective of the ADMIRE project was to establish control by creating an active I/O stack that dynamically adjusts computation and storage requirements through intelligent global coordination, malleability of computation and I/O, and the scheduling of storage resources along all levels of the storage hierarchy, via a software-defined framework based on scalable monitoring and control, separated control and data paths, and orchestration of key system components and applications through embedded control points. I was responsible for WP6 (Intelligent Controller).
• With Pete Beckman and Swann Perarnau from ANL we organized the RADR workshop (in conjunction with IPDPS).
• I was the program chair of the COLOC workshop (in conjunction with Euro-Par).
• I was the team leader of the TADaaM Inria research project (2015-2024). The goal of the TADaaM Inria project (joint with Univ. Bordeaux and Bordeaux INP) is to design and build a stateful system-wide service layer for HPC systems. This layer abstracts low-level features of the system (e.g. topology, network, resource usage) and the software stack (e.g. threads, data), and lets applications register their needs and behavior via a carefully designed API, optimizing the execution of all running applications in a coordinated fashion and at system scale.
• The GRID'5000 project aims at building a highly reconfigurable, controllable and monitorable experimental Grid platform gathering 9 sites geographically distributed in France, featuring a total of 5000 CPUs. I was responsible for the Nancy site (3 clusters, 96/240/184 CPUs) and the Bordeaux site (2 clusters).
• I was member of the Inria project Lab on the convergence of HPC and Big Data (2018-2021).
• I was member of the steering committee of the Euro-Par conference from 2011 to 2021.
• I was member of the steering committee of the Cluster conference from 2010 to 2020.
• I was the vice chair of the Network for Sustainable Ultrascale Computing (NESUS) European COST Action IC 1305. NESUS established an open European research network targeting sustainable solutions for ultrascale computing, cross-fertilizing HPC, large-scale distributed systems, and big data management (2013-2018).
• I was member and chair of the Inria partner of the ELCI project (Environnement Logiciel pour le Calcul Intensif / Software Environment for High-Performance Computing), which aimed to improve numerical-simulation and computer science techniques so that real-world simulation codes fully exploit future generations of high-end hardware.
• I was member of the COLOC ITEA project, providing new models, mechanisms and tools for improving application performance and supercomputer resource usage while taking into account data locality and concurrency. I was responsible for WP2: data transfer and locality modeling.
• I was member of the MOEBUS ANR project, focused on the efficient execution of parallel applications submitted by various users sharing resources in large-scale HPC environments. I was head of task 3: Multi-objectives Scheduling. Duration: 4 years (2013-2017).
• I was member of the MORSE associated team between INRIA, UTK and UCD. The goal of Matrices Over Runtime Systems at Exascale (MORSE) was to design dense and sparse linear algebra methods that achieve the fastest possible time to an accurate solution on large-scale multicore systems with GPU accelerators.
• I was member of the Inria Runtime team until 2015.
• I led the European COST Action ComplexHPC: Open European Network for High-Performance Computing in Complex Environments from 2009 to 2013, gathering more than 20 countries and 30 partners in Europe, focused on high-performance heterogeneous computing.
• I was the leader of the ALEAE INRIA ARC (2009-2011), providing models and algorithmic solutions for resource management under uncertainty in large-scale distributed systems.
• The goal of wrekavoc is to define and control the heterogeneity of a given platform by degrading CPU, network or memory capabilities of each node composing this platform.
• Network of excellence CoreGRID, a European research network on foundations, software infrastructures and applications for large-scale distributed, grid and peer-to-peer technologies. Member of work-package 6 "resources management" (2005-2008).
• ACI data masses AGIR: new image processing algorithms on the GRID, progressive transmission of medical images.
• NSF-INRIA project, collaboration with ICL of the Univ. of Tennessee and the GRAAL project of INRIA. Collaboration with the Netsolve team on adaptive online compression.
• ARC INRIA OTaPHe: scheduling parallel tasks on heterogeneous platforms.
• ACI data masses GRID-Explorer: design of a GRID emulator. Responsible for the heterogeneity emulation task.
• ARC INRIA RedGRID: data redistribution for code-coupling applications and computational steering.
• I collaborated on the DIET environment, a PSE with a hierarchical set of agents.
• ACI GRID ASP: collaboration with Gerald Monard from ECBT. I designed an application for computing the potential energy surface of a system of molecules.
• During my postdoc I worked within the INRIA ARC OURAGAN on Scilab parallel, which aims to provide transparent access to distributed resources, e.g. interfacing Scilab with Netsolve.
• Euro-tops project with Matra-Caps systems: automatic parallelization of high-performance applications.

Past events

• With Guillaume Pallez I organized ICPP 22 (51st International Conference on Parallel Processing) in Bordeaux.
• I was member of the steering committee of the IEEE cluster conference (2010-2020).
• I was track chair (area: Application, Algorithms, and Libraries) of the 2020 IEEE cluster conference.
• I was publicity chair (for Europe) of the 2021 IEEE cluster conference.
• I was one of the organizers of the Fifth Workshop on Programming Abstractions for Data Locality (PADAL'19) in Bordeaux (Sept. 2019).
• I was program chair of HCW 2017 (in conjunction with IPDPS) and the COLOC workshop 2017 (in conjunction with EuroPar).
• With Alexandre Denis, Brice Goglin and Guillaume Mercier, I was member of the organizing committee of Euro-MPI 2015.
• I was program chair of Heteropar 2014.
• Euro-Par is a leading conference in the field of parallelism and distributed computing. I was main organizer and vice chair of the Euro-Par 2011 edition.
• I participated in organizing the 2009 GRID'5000 Spring school, which took place in Nancy in April 2009.
• I participated in organizing the GridUSe summer school at Supelec Metz in 2004.
• I organized the winter school GRID'2002, which took place in Aussois in December 2002.