Section: New Software and Platforms
Keywords: HPC - Cloud - Clusters - Resource manager - Light grid
Scientific Description This batch system is based on a database (PostgreSQL (preferred) or MySQL), a script language (Perl) and an optional scalable administrative tool (e.g. Taktuk). It is composed of modules which interact mainly via the database and are executed as independent programs. Therefore, formally, there is no API, the system interaction is completely defined by the database schema. This approach eases the development of specific modules. Indeed, each module (such as schedulers) may be developed in any language having a database access library.
Functional Description OAR is a versatile resource and task manager (also called a batch scheduler) for HPC clusters, and other computing infrastructures (like distributed computing experimental testbeds where versatility is a key).
The OAR ecosystem also include several associated software tools that proved to be useful independently from OAR. Among theses, two softwares play a major role in the support our research studies. The first one is Kameleon (http://kameleon.imag.fr), a tool to help enhancing reproducibility of experiments by guarantee the ability to reproduce the complete used software stacks. The second one is Batsim (https://gforge.inria.fr/projects/batsim) a RJMS simulator based on SimGrid. Batsim simulates job execution taking into account the target platform hardware capabilities through SimGrid, while scheduling is performed by an actual job management system. A comprehensive API enables to easily plug into BatSim various job management systems like OAR.