Methodology for the Implementation of Slurm-based HPC Services

Nowadays, with the growth of the increasing demand for computational capabilities, it is necessary to have workstations with extreme computational capacity, for which we find in the market many solutions, depending on the needs and funding, in most cases commercial solutions, have high costs. As an alternative we present a methodology for the implementation of HPC services, using the SLURM tool, which provides mechanisms to perform HPC processes at low cost, with the resources that are available, we present the requirements and procedures for the configuration of these solutions, as well as the demonstration of operation, finally we indicate as conclusions, the mechanisms for the application and implementation of the methodology in different cases, where it is required to perform high performance processes based on HPC solutions.