EPJ Web of Conferences 245, 07006 (2020)
https://doi.org/10.1051/epjconf/202024507006

ComputeOps: Container for High Performance Computing

¹ APC/FACe, Université de Paris, CNRS/IN2P3, CEA/Irfu, Obs. de Paris, 10 rue A. Domon et L. Duquet, 75013 Paris, France
² Centre de Calcul de l’IN2P3, 21 Avenue Pierre de Coubertin, 69100 Villeurbanne, France
³ LLR, Ecole Polytechnique, Rue de Fresnel, 91128 Palaiseau, France
⁴ ISCD, BP380, 4, place Jussieu, 75252 Paris Cedex 5, France
⁵ LPNHE, Campus de Jussieu, 4 place Jussieu, 75252 Paris Cedex 5, France
⁶ IJCLab, CNRS Université Paris-Saclay Bât. 100, Faculté des sciences, F-91405 Orsay Cedex
⁷ IPHC, 23, rue du Loess BP28, 67037 Strasbourg Cedex 2, France

^* e-mail: ccavet@apc.in2p3.fr
^** e-mail: souchal@apc.in2p3.fr
^*** e-mail: sebastien.gadrat@cc.in2p3.fr

Published online: 16 November 2020

Abstract

The High Performance Computing (HPC) domain aims to optimize code in order to use the latest multicore and parallel technologies including specific processor instructions. In this computing framework, portability and reproducibility are key concepts. A way to handle these requirements is to use Linux containers. These “light virtual machines” allow to encapsulate applications within its environment in Linux processes. Containers have been recently rediscovered due to their abilities to provide both multi-infrastructure environnement for developers and system administrators and reproducibility due to image building file. Two container solutions are emerging: Docker for microservices and Singularity for computing applications. We present here the status of the ComputeOps project which has the goal to study the benefit of containers for HPC applications.