Photo

François Tessier

Phone: (+33) [TBA]
Email:
[GScholar] Google Scholar

Inria

I have been a Research Scientist at Inria in Rennes, within the KerData team since November 2020. I have been focusing there on I/O and storage challenges in a context of HPC/Cloud convergence. Before that, I stayed two years as a Computational Scientist at ETH Zürich within CSCS (Swiss National Supercomputing Center) located in Lugano, Switzerland where worked on dynamic resource provisioning for HPC applications and large-scale workflows. Previously, I spent two years and a half as a postdoctoral appointee at Argonne National Laboratory, IL, USA where I addressed the problem I/O optimization. I received a Ph.D in Computer Science in 2015 from University of Bordeaux, France, under the supervision of Emmanuel Jeannot and Guillaume Mercier. My Ph.D thesis focused on topology and affinity-aware application placement and load balancing.

My research interests are mainly around I/O optimization and the efficient use of storage resources on large-scale systems. I am also interested in topology-aware placement techniques.

Research

Publications [GScholar]

[ABC] Authors by alphabetical order
Filter by year:

International Journals


  • François Tessier, Venkatram Vishwanath, Emmanuel Jeannot - Adding Topology and Memory Awareness in Data Aggregation Algorithms - Future Generation Computer Systems (FGCS), May 2024 - [WWW] - Paper
  • Julien Monniot, François Tessier, Matthieu Robert, Gabriel Antoniu - Supporting dynamic allocation of heterogeneous storage resources on HPC systems - Concurrency and Computation, Practice and Experience (CCPE), Sept 2023 - [WWW] - Paper
  • [ABC] Emmanuel Jeannot, Guillaume Mercier and François Tessier - Process Placement in Multicore Clusters: Algorithmic Issues and Practical Techniques - IEEE Transactions on Parallel and Distributed Systems (TPDS), April 2014 - [WWW] - Paper - Appendix - [BibTeX]

Reviewed International Conferences with proceedings


  • Julien Monniot, François Tessier, Henri Casanova, Gabriel Antoniu - Simulation of Large-Scale HPC Storage Systems: Challenges and Methodologies - HiPC 2024, 31st IEEE International Conference on High Performance Computing, Data, and Analytics, Bangalore, India (December 2024) - [WWW] - Paper
  • Théo Jolivel, François Tessier, Julien Monniot, Guillaume Pallez - MOSAIC: Detection and Categorization of I/O Patterns in HPC Applications - PDSW 2024, 9th International Parallel Data Systems Workshop, Held in conjunction with ACM/IEEE SuperComputing'24 Conference, Atlanta, GA, USA (November 2024) - [WWW] - Paper
  • Julien Monniot, François Tessier, Matthieu Robert, Gabriel Antoniu - StorAlloc: A Simulator for Job Scheduling on Heterogeneous Storage Resources - HeteroPar 2022, 20th International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Platforms, Glasgow, Scotland, UK (August 2022) - [WWW] - Paper
  • Christopher Haine, Utz-Uwe Haus, François Tessier, Domokos Sármány, Maxime Martinasso, Tiago Quintino, Dirk Pleiter, Simon Smart, Adrian Tate - A Middleware Supporting Data Movement in Complex and Software-Defined Storage and Memory Architectures - ISC High Performance 2021, 4th Workshop on Interoperability of Supercomputing and Cloud Technologies, Frankfurt, Germany (Online, July 2021) - [WWW] - Paper
  • François Tessier, Maxime Martinasso, Matteo Chesi, Mark Klein, Miguel Gila - Dynamic Provisioning of Storage Resources: a Case Study with Burst Buffers - 2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), New Orleans, LA (May 2020) - [WWW] - Paper - Talk - [BibTeX]
  • François Tessier, Paul Gressier, Venkatram Vishwanath - Optimizing Data Aggregation by Leveraging the Deep Memory Hierarchy on Large-Scale Systems - ICS 2018: The 32nd ACM International Conference on Supercomputing, Beijing, China (June 2018) (Double-blind, Acceptance rate: 18.7%) - [WWW] - Paper - Talk - [BibTeX]
  • Houjun Tang, Suren Byna, François Tessier, Teng Wang, Bin Dong, Jingqing Mu, Quincey Koziol, Jerome Soumagne, Venkatram Vishwanath, Jialin Liu, Richard Warren - Toward Scalable and Asynchronous Object-centric Data Management for HPC - 18th IEEE/ACM CCGrid 2018, Washington, DC (May 2018) (Acceptance rate: 20.8%) - [WWW] - Paper - Talk - [BibTeX]
  • François Tessier, Venkatram Vishwanath, Emmanuel Jeannot - TAPIOCA: An I/O Library for Optimized Topology-Aware Data Aggregation on Large-Scale Supercomputers - IEEE Cluster 2017, Honolulu, HI (Sept. 2017) (Acceptance rate: 21.8% - Area: 15%) - [WWW] - Paper - Talk - [BibTeX]
  • Jialin Liu, Quincey Koziol, Houjun Tang, François Tessier, Wahid Bhimji, Brandon Cook, Brian Austin, Suren Byna, Bhupender Thakur, Glenn Lockwood, Jack Deslippe, Prabhat - Understanding the I/O Performance Gap Between Cori KNL and Haswell - Cray User Group Conference 2017 (CUG 2017), Redmond, WA (2017/05/10) - [WWW] - Paper - Talk - [BibTeX]
  • Thomas Marrinan, Joseph A. Insley, Silvio Rizzi, François Tessier and Michael E. Papka - Automated Dynamic Data Redistribution - 18th IEEE International Workshop on Parallel and Distributed Scientific and Engineering Computing (PDSEC 2017), Held in conjunction with the 31st IEEE IPDPS'17 Conference, Orlando, FL (2017/06/02) - [WWW] - Paper - [BibTeX]
  • François Tessier, Preeti Malakar, Venkatram Vishwanath, Emmanuel Jeannot and Florin Isaila - Topology-Aware Data Aggregation for Intensive I/O on Large-Scale Supercomputers - 1st Workshop on Optimization of Communication in HPC runtime systems (IEEE COM-HPC16), Held in conjunction with ACM/IEEE SuperComputing'16 Conference, Salt Lake City, UT (2016/11/18) (Acceptance rate: 35%) - [WWW] - Paper - Talk - [BibTeX]
  • [ABC] Emmanuel Jeannot, Guillaume Mercier and François Tessier - Topology and Affinity Aware Hierarchical and Distributed Load-balancing in Charm++ - 1st Workshop on Optimization of Communication in HPC runtime systems (IEEE COM-HPC16), Held in conjunction with ACM/IEEE SuperComputing'16 Conference, Salt Lake City, UT (2016/11/18) (Acceptance rate: 35%) - [WWW] - Paper - Talk - [BibTeX]
  • [ABC] Emmanuel Jeannot, Esteban Meneses, Guillaume Mercier, François Tessier and Gengbin Zheng - Communication and Topology-aware Load Balancing in Charm++ with TreeMatch - IEEE Cluster 2013, Indianapolis, IN (2013/09/24) - [WWW] - Paper - Talk - [BibTeX]

Reviewed International Conferences without proceedings


  • François Tessier, Maxime Martinasso, Matteo Chesi, Mark Klein, Miguel Gila - Dynamically Provisioning Cray DataWarp Storage - Cray User Group Conference 2019 (CUG 2019), Montréal, Canada (May 2019) - [WWW] - [arXiv] - Paper - Talk
  • Suren Byna, Quincey Koziol, Venkatram Vishwanath, Jerome Soumagne, Houjun Tang, Jingqing Mu, Bin Dong, Richard Warren, François Tessier, Teng Wang, Jialin Liu - Proactive Data Containers (PDC): An Object-centric Data Store for Large-scale Computing Systems - 2018 AGU Fall Meeting, Washington, D.C. (2018/12/12) - [WWW] - [TBA]
  • François Tessier, Paul Coffman, Preeti Malakar, Richard Zamora, Venkatram Vishwanath and Georges Brown - Empirical Study of I/O Performance of Lustre-based Storage System on a Cray XC40 Supercomputer - 2018 Lustre User Group (LUG) Conference, Chicago, IL (Apr. 2018) - [WWW] - Talk
  • [ABC] Emmanuel Jeannot, Guillaume Mercier and François Tessier- Matching communication pattern with underlying hardware architecture - 6th European Conference on Computational Fluid Dynamics, Barcelona, Spain (Jul. 2014) - Paper - [BibTeX]

Reviewed National Conferences


  • [ABC] Emmanuel Jeannot, Guillaume Mercier and François Tessier - TreeMatch : Un algorithme de placement de processus sur architectures multicœurs - ComPAS' 2013 - RenPar, Grenoble (2013/01/17) - [WWW] - Paper - Talk - [BibTeX]

Posters


  • Julien Monniot, François Tessier, Gabriel Antoniu - Modeling Allocation of Heterogeneous Storage Resources on HPC Systems - Supercomputing'22, Dallas, TX (Nov. 2022) (Best Research Poster Award nominee, 4 posters selected out of ~90) - Poster
  • Maestro Consortium - Middleware for Memory and Data-Awareness in Workflows - ISC 2019, Frankfurt, Germany (June 2019) - Poster
  • Sergio Servantez, Richard Zamora, François Tessier, Zhiling Lan - Using RAM Area Network to Reduce Synchronization Costs in Collective I/O Operations - SIAM Conference on Computational Science and Engineering 2019, Spokane, WA, USA (Feb. 2019) - Poster
  • François Tessier, Preeti Malakar, Venkatram Vishwanath, Emmanuel Jeannot, Florin Isaila - Topology-aware data aggregation for parallel I/O on Blue Gene/Q - GCASR Workshop 2016, Chicago, IL (2016/04/27) - Poster

White Papers, Research Reports and Preprints


  • Joshua Bowden, François Tessier, Charles Deltel, Simone Bnà, Gabriel Antoniu - In-situ visualization using Damaris: The Code_Saturne use case - PRACE White Paper (2021) - [WWW] - Paper
  • Jerome Soumagne, Richard Warren, Jingqing Mu, Neelam Bagha, Venkat Vishwanath, Francois Tessier, Suren Byna, Quincey Koziol, Houjun Tang, Teng Wang, Bin Dong, and Jialin Liu - Proactive Data Containers for Scientific Storage (Final Report) - US DoE Final Technical Report - [OSTI]
  • George K. Thiruvathukal, Cameron Christensen, Xiaoyong Jin, François Tessier, Venkatram Vishwanath - A Benchmarking Study to Evaluate Apache Spark on Large-Scale Supercomputers - ArXiv preprint (Apr. 2019) - [arXiv] - Paper - [BibTeX]
  • Sergio Servantez, Rick Zamora, François Tessier, Zhiling Lan - Collective I/O using RAM Area Network - Project Report, Illinois Institute of Technology / Argonne National Laboratory (Mar. 2019) - Paper
  • François Tessier, Venkatram Vishwanath - Reproducibility and Variability of I/O Performance on BG/Q: Lessons Learned from a Data Aggregation Algorithm - Research Report, Argonne National Laboratory (Dec. 2017) - Paper - [BibTeX]
  • [ABC] Emmanuel Jeannot, Guillaume Mercier and François Tessier - Process Placement in Multicore Clusters: Algorithmic Issues and Practical Techniques - Inria Research Report, May 2013 - [WWW] - Inria Research Report N°8269 - [BibTeX]

PhD. Thesis


  • François Tessier - Placement d'applications parallèles en fonction de l'affinité et de la topologie - Ph.D. Thesis, French (2015/01/26) - [WWW] - Paper - Talk - [BibTeX]

Talks

  • [Invited Talk] Maestro Project Introduction - Fifth Workshop on Programming Abstractions for Data Locality (PADAL'19), Bordeaux, France (2019/09/09) - Talk
  • [Paul Coffman] Parallel I/O on Theta with Best Practices - Argonne National Lab. SDL Workshop, Lemont, IL (2018/03/01) - Talk
  • [Invited Talk] TAPIOCA: An I/O Library for Optimized Topology-Aware Data Aggregation on Large-Scale Supercomputers - Seminar at Inria Rhône-Alpes, Grenoble (2017/12/21) - Talk
  • Toward portable I/O performance by leveraging system abstractions of deep memory and interconnect hierarchies - US Department of Energy Centers of Excellence - Performance Portability Meeting 2017, Denver, CO (2017/08/23) - Talk
  • [Invited Talk] System abstractions to facilitate data movement in supercomputers with deep memory and interconnect hierarchy - PADAL'17 Workshop, Chicago, IL (2017/08/03) - Talk
  • Short Talk: System abstractions to facilitate data movement in supercomputers with deep memory and interconnect hierarchy - 7th JLESC Workshop, UIUC, Urbana-Champaign, IL (2017/07/17) - Talk
  • Toward taming large and complex data flows in data-centric supercomputing - 6th JLESC Workshop, RIKEN AICS, Kobe, JAPAN (2016/12/01) - Talk
  • Distributed communication-aware load balancing with TreeMatch in Charm++ - 9th Scheduling for Large Scale Systems Workshop, ENS Lyon (2014/07/01) - Talk
  • Distributed communication-aware load balancing with TreeMatch in Charm++ - 11th JLPC Workshop (1st JLESC Workshop), Sophia-Antipolis (2014/06/09) - Talk
  • Charm++, what’s that?! - Les Mardis du dev' - Internal training day (2013/10/15) - Talk
  • Communication-aware load balancing with TreeMatch in Charm++ - 9th JLPC Workshop, Lyon (2013/06/14) - Talk
  • Load balacing and affinities between processes with TreeMatch in Charm++: preliminary results and prospects - 7th JLPC Workshop, Rennes (2012/06/15) - Talk
  • Processes placement on multicore. Dynamic load balancing in Charm++ - 10th Charm++ Workshop, Urbana-Champaign, IL (2012/05/07) - Talk

Major Software Developments

  • Intermediate Storage As A Cache (ISAAC) : Lightweight library implementing a wrapper of standard MPI-IO calls. Targeting supercomputers with node-local storage (HDD, SSD), ISAAC can take advantage of this memory layer and use it as a transparent cache to significantly improve the I/O bandwidth.
  • TAPIOCA : Topology-aware data aggregation C++ library for I/O intensive applications.
  • TreeMatchLB : Set of topology-aware load balancing algorithms developed for the Charm++ runtime system including TMLB_TreeBased, a highly-scalable hierarchical and distributed load balancer targetting communication-bound applications.
  • TreeMatch : Algorithm and tool to find a good process placement according to the communication pattern of an application and the target architecture. Project initially developed by Emmanuel Jeannot and Guillaume Mercier.

Collaborations

  • Proactive Data Containers [WWW]
    • Data movement optimizations
      • Meeting: Lawrence Berkeley National Laboratory - Berkeley, CA - 2017/04/19 to 2017/04/20
      • Remote weekly meeting since March 2016
  • Joint Laboratory for Extreme Scale Computing [WWW]
    • Toward taming large and complex data flows in data-centric supercomputing
  • Joint Laboratory for Petascale Computing [WWW]
    • NCSA - Urbana-Champaign, IL : Listener of the 8th JLPC Workshop
    • Visit: Parallel Programming Laboratory - Urbana-Champaign, IL : Two weeks at the beginning of May 2012 (Charm++ Workshop, affinity-aware load balancing)
    • Visit: Parallel Programming Laboratory - Urbana-Champaign, IL : One week at the end of August 2013 (Distributed and parallel affinity-aware load balancing)

Responsibilities

Students Supervision

  • Sergio Servantez (Undergraduate, Illinois Institute of Technology, since April 2018): Using RAM Area Network to Reduce Synchronization Costs in Collective I/O Operations
  • Paul Gressier (Undergraduate, ENSEIRB Engineering school, 3 months internship in 2017): Optimizing Data Movement with Data Transformations

Scientific Mediation

  • La Théorie des Jeux (Game Theory) - Workshop for the "Fête de la Science" at Inria Bordeaux - 2015
  • SC14 at New Orleans, LA (Nov. 2014) : Demonstration of the 3d-printed and free humanoid robot Poppy for the Inria booth.
  • Presentation of a part of my thesis during the poster session of the EDMI day (2014/11/05) - Poster
  • Découvrez le parallélisme en faisant des gauffres (Discover parallelism by making waffles) - Workshop for the "Fête de la Science" at Inria Bordeaux - 2013
  • Exhibitor for the Inria booth at Aquitec, an annual exhibition of Training and Carrier - 2013

Teaching (French)

2013-2014 : IUT Informatique Bordeaux

2012-2013 : ENSEIRB

Miscellaneous

  • Vice-president of the associative DIY ISP Aquilenet
  • Board's member of the associative DIY ISPs federation FFDN