Výsledky vyhledávání - "Le Fèvre, Valentin"

Comparing Distributed Termination Detection Algorithms for Task-Based Runtime Systems on HPC platforms

Autor: Bosilca, George, Bouteiller, Aurélien, Hérault, Thomas, Le Fèvre, Valentin, Robert, Yves, Dongarra, Jack

Publikováno v: International Journal of Networking and Computing
International Journal of Networking and Computing, 2022, 12 (1)

International audience; This paper revisits distributed termination detection algorithms in the context of High-Performance Computing (HPC) applications. We introduce an efficient variant of the Credit Distribution Algorithm (CDA) and compare it to t

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::5bac12d1175beb84e26600308b368e4f
https://inria.hal.science/hal-03920388/file/ijnc22.pdf

Zobrazit plný text záznamu

Ordonnancement avec tolérance aux pannes pour des tâches parallèles à nombre de processeurs programmable

Autor: Benoit, Anne, Le Fèvre, Valentin, Perotin, Lucas, Raghavan, Padma, Robert, Yves, Sun, Hongyang

Publikováno v: [Research Report] RR-9340, Inria-Research Centre Grenoble – Rhône-Alpes. 2021

We study the resilient scheduling of moldable parallel jobs on high-performance computing (HPC) platforms. Moldable jobs allow for choosing a processor allocation before execution, and their execution time obeys various speedup models. The objective

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______3393::ab93c7472648ec65a4d360364a333ee1
https://inria.hal.science/hal-02614215

Zobrazit plný text záznamu

Resilient scheduling algorithms for large-scale platforms

Autor: Le Fèvre, Valentin

Publikováno v: Distributed, Parallel, and Cluster Computing [cs.DC]. Université de Lyon, 2020. English. ⟨NNT : 2020LYSEN019⟩

This thesis focuses on a major problem for the HPC community: resilience. Computing platforms are bigger and bigger in order to reach what we call exascale, i.e. a computing capacity of 10^18 FLOP/s but they suffer numerous failures. Reducing the exe

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od_______212::9ed0c9db7e3c758dc65da4d573159a90
https://tel.archives-ouvertes.fr/tel-02947051/file/LE_FEVRE_Valentin_2020LYSEN019_These.pdf

Zobrazit plný text záznamu

Ordonnancement avec tolérance aux pannes pour des tâches parallèles à nombre de processeurs programmable

Autor: Benoit, Anne, Le Fèvre, Valentin, Perotin, Lucas, Raghavan, Padma, Robert, Yves, Sun, Hongyang

Publikováno v: [Research Report] RR-9340, Inria-Research Centre Grenoble – Rhône-Alpes. 2020
[Research Report] RR-9340, Inria-Research Centre Grenoble – Rhône-Alpes. 2021

This paper focuses on the resilient scheduling of moldable parallel jobson high-performance computing (HPC) platforms. Moldable jobs allow for choosing aprocessor allocation before execution, and their execution time obeys various speedup models. The

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::250aea526190e8a3dff8aa9082926af1
https://hal.inria.fr/hal-02614215

Zobrazit plný text záznamu

Combining Checkpointing and Replication for Reliable Execution of Linear Workflows with Fail-Stop and Silent Errors

Autor: Benoit, Anne, Cavelan, Aurélien, Ciorba, Florina, Le Fèvre, Valentin, Robert, Yves

Publikováno v: [Research Report] RR-9235, ROMA (INRIA Rhône-Alpes / LIP Laboratoire de l’Informatique du Parallélisme); LIP-Laboratoire de l’Informatique du Parallélisme. 2018, pp.1-32

Large-scale platforms currently experience errors from two different sources,namely fail-stop errors (which interrupt the execution) and silent errors (which strikeunnoticed and corrupt data). This work combines checkpointing and replication for ther

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od_______212::25c1e9c9818d7b705d5ebb4bc32fdbfa
https://hal.inria.fr/hal-01955859

Zobrazit plný text záznamu

Distributed Termination Detection for HPC Task-Based Environments

Autor: Bosilca, George, Bouteiller, Aurelien, Hérault, Thomas, Le Fèvre, Valentin, Robert, Yves, Dongarra, Jack

Publikováno v: [Research Report] RR-9181, Inria-Research Centre Grenoble – Rhône-Alpes. 2018, pp.1-28
[Research Report] RR-9181, Inria-Research Centre Grenoble – Rhône-Alpes. 2018, pp.28

This paper revisits distributed termination detection algorithms in the context of high-performance computing applications in task systems. We first outline the need to efficiently detect termination in workflows for which the total number of tasks i

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::f5c3e0f955b99ce870e943200b003fbc
https://inria.hal.science/hal-01811823

Zobrazit plný text záznamu

Combining Checkpointing and Replication for Reliable Execution of Linear Workflows

Autor: Benoit, Anne, Cavelan, Aurélien, Ciorba, Florina, Le Fèvre, Valentin, Robert, Yves

Publikováno v: [Research Report] RR-9152, Inria-Research Centre Grenoble – Rhône-Alpes. 2018, pp.1-36

This report combines checkpointing and replication for the reliable executionof linear workows. While both methods have been studied separately, their combinationhas not yet been investigated despite its promising potential to minimize the execution

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od_______212::4053fc0e3d1644dddbbcfca8328eea59
https://hal.inria.fr/hal-01714978

Zobrazit plný text záznamu

Akademický článek

Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.

Vyhledávací nástroje:

Upřesnit hledání