Publications

Thèse

• Lambert, Thomas. Étude de l'effet de la réplication de fichiers d'entrée sur l'efficacité et la robustesse d'un ensemble de calculs. Manuscrit de thèse. Université de Bordeaux, France, 2017.

Journaux Internationaux

• [FGCS 2020] Olivier Beaumont, Thomas Lambert, Loris Marchal, Bastien Thomas. Performance Analysis and Optimality Results for Data-Locality Aware Tasks Scheduling with Replicated Inputs. In Future Generation Computer Systems, 111: p.582-598, 2020.
  
• [TPDS 2018] Olivier Beaumont, Brett A. Becker, Ashley DeFlumere, Lionel Eyraud-Dubois, Thomas Lambert and Alexey Lastovetsky. Recent Advances in Matrix Partitioning for Parallel Computing on Heterogeneous Platforms. In Transactions on Parallel and Distributed Systems, 30.1: p.218-229, 2018.
  
• [TCS 2018] Enver Kayaaslan, Thomas Lambert, Loris Marchal, and Bora Uçar. Scheduling Series-Parallel Task Graphs to Minimize Peak Memory. In Theoretical Computer Science, 707: p.1-23, 2018.
  
• [IPL 2015] Thomas Lambert, Loris Marchal, Bora Uçar, Comments on the hierarchically structured bin packing problem. In Information Processing Letters, 115(2): p.306-309, 2015.
  

Conférences Internationales

• [ICCCN 2019] Jad Darrous, Thomas Lambert, Shadi Ibrahim. On the Importance of Container Image Placement for Service Provisioning in the Edge. In 28th International Conference on Computer Communications and Networks, p.1-9, 2019.
  
• [ICPP 2018] Lionel Eyraud-Dubois and Thomas Lambert. Using Static Allocation Algorithms for Matrix Matrix Multiplication on Multicores and GPUs. In International Conference on Parallel Processing, p.1-10, 2018.
  
• [Euro-Par 2016] Olivier Beaumont, Lionel Eyraud-Dubois, Thomas Lambert Cuboid Partitioning for Parallel Matrix Multiplication on Heterogeneous Platforms. In International European Conference on Parallel and Distributed Computing, p.171-182, 2016.
  
• [IPDPS 2016] Olivier Beaumont, Lionel Eyraud-Dubois, Thomas Lambert A New Approximation Algorithm for Matrix Partitioning in Presence of Strongly Heterogeneous Processors. In International Parallel and Distributed Processing Symposium, p.474-483, 2016.
  
• [SBAC-PAD 2015] Olivier Beaumont, Lionel Eyraud-Dubois, Abdou Guermouche, Thomas Lambert Comparison of Static and Dynamic Resource Allocation Strategies for Matrix Multiplication. In International Symposium on Computer Architecture and High Performance Computing, p.170-177, 2015.
  
• [COCOON 2013] Christophe Crespelle, Eric Thierry, Thomas Lambert, A Linear-Time Algorithm for Computing the Prime Decomposition of a Directed Graph with Regard to the Cartesian Product. In 19th Annual International Computing and Combinatorics Conference, p.469-480, 2013.
  

Workshops et Short Papers

• [CIKM 2020] Thomas Lambert, David Guyon and Shadi Ibrahim. Rethinking Operators Placement of Stream Data Application in the Edge. In 29th ACM International Conference on Information & Knowledge Management, p.2101-2104, 2020.
  
• [Cloud Workshops 2018] Olivier Beaumont, Thomas Lambert, Loris Marchal, and Bastien Thomas. Data-Locality Aware Dynamic Schedulers for Independent Tasks with Replicated Inputs. In IEEE Cloud Workshops, p.516-523, 2018.
  
• [CEBDA 2018] Thomas Lambert and Rizos Sakellariou. Allocation of a Publisher/Subscriber Model Links on a Set of Virtual Machines. In International workshop on the Convergence of Extreme Scale Computing and Big Data Analysis, p.1206-1213, 2018.
  

Code

• Une implémentation du produit du matrices pour architecture hétérogène (avec la librairie StarPU) et de plusieurs algorithmes de partitionnement de carré ou cube sont disponible ici. Ce code a été utilisé pour notre publication dans ICPP 2018. N'hésitez pas à me demander si vous avez la moindre question.