题 目：The Wolphin Industrial Project: Scheduling Containers in a Cloud and more
报告人：Christophe Cérin 法国巴黎十三大学
This presentation is related to the Wolphin joint R&D effort between French labs (LIP6, LIPN) and companies (Alter Way, Objectif Libre, Gandi) to tackle the management and scheduling optimization problems in an unknown context of containers based infrastructures. We showcase research and industrial strategies to optimise resource management for elastic loads, with offline and online demands and model-based approaches. After an overview of the project and contributions in general, we make a focus on the scheduling of containers problem according to a new economic model. The constraints on the number of required resources (RAM, CPU, DISK) for the execution is relaxed and this provides with opportunities, not only for the scheduling problem, but also for many other problems (consolidation). We show experimental works and draw perspectives regarding the broad range of problems with containers, on Clouds but also on HPC systems.
Christophe Cérin has been a professor of computer science at the University of Paris 13, France since 2005. In 2015, he was involved with an infrastructure project related to big data and high performance computing for e-sciences for USPC (Université Sorbonne Paris Cité). At Paris13, he chairs the board for the cluster computing facility available to all campus scientists and he chaired (2011-2016) the 'Expert Committee' in charge of recruiting and mentoring full time, junior and senior professors in computer science. His current industrial experience includes serving as local chair for the Wolphin project (Alterway, Gandhi, Objectif Libre and Paris 6). His recent industrial experience was for the Wendelin and Resilience projects related to Cloud and Big-Data. He is also active with RSU (industrial demonstrator for sustainable cities), with Qarnot Computing and Umanis. He is currently an invited professor (2019-2023) at Hangzhou Dianzi University, China. His research focuses on High Performance Computing, including Grid and Cloud Computing and he develops middleware, algorithms, tools and methods for distributed systems.