Batch System Prototype Using Containers

The successfully operated data center of the Meshcheryakov Laboratory of Information Technologies (MLIT, JINR) experiences a continuous demand for new computational resources, both from the experiments of the NICA accelerator complex and from facilities at the LHC (CERN). The SLURM system is used for batch data processing. In traditional practice, task execution takes place on a physical server (compute node) with an installed and configured operating system, as well as middleware. The article considers the implemented prototype of a batch system using a containerized environment. The use of containers can allow for the utilization of more than one type of resource for task execution and, minimize operational efforts for preparing compute nodes.