Thomas Bouvier

4
Documents

Présentation

My main current research interests are related to deployment and scheduling strategies for AI workloads on heterogeneous resources (HPC, cloud, fog, edge).

Relevant topics:

  • Cloud data services and big data analytics
  • Stream processing
  • Parallel and distributed deep learning
  • Online learning
  • Workflow management and placement strategies

Publications

Publications

Image document

Efficient Data-Parallel Continual Learning with Asynchronous Distributed Rehearsal Buffers

Thomas Bouvier , Bogdan Nicolae , Hugo Chaugier , Alexandru Costan , Ian Foster et al.
CCGrid 2024 - IEEE 24th International Symposium on Cluster, Cloud and Internet Computing, May 2024, Philadelphia (PA), United States. pp.1-10, ⟨10.1109/CCGrid59990.2024.00036⟩
Communication dans un congrès hal-04600107 v1
Image document

Deploying Heterogeneity-aware Deep Learning Workloads on the Computing Continuum

Thomas Bouvier , Alexandru Costan , Gabriel Antoniu
BDA 2021 - 37e Conférence sur la Gestion de Données - Principes, Technologies et Applications, Oct 2021, Paris, France
Communication dans un congrès hal-03338520 v2
Image document

Heterogeneity-aware Deep Learning Workload Deployments on the Computing Continuum

Thomas Bouvier , Alexandru Costan , Gabriel Antoniu
IPDPS 2021 - 35th IEEE International Parallel and Distributed Processing Symposium, May 2021, Virtual / Portland, United States
Poster de conférence hal-03270129 v1