BSC-CNS
Published on BSC-CNS (https://www.bsc.es)

Inici > Dynamic resource management in HPC

Dynamic resource management in HPC

Dynamic resource allocation and scheduling is key for an optimal resource utilization. In this research line we target system software for a smart allocation taking into consideration application characteristics for MPI+OpenMP applications. 

Summary

Dynamic resource allocation and scheduling is key for an optimal resource utilization. In this research line we target system software for a smart allocation taking into consideration application characteristics for MPI+OpenMP applications. Our research uses SLURM as DRM as a proff of concept of our proposals in real systems and the BSC SLURM simulator  as the platform for large scale experiments. 

Objectives

  • Dynamic resource allocation in HPC
    • Coodination between applications and scheduler
    • Holistic job scheduling for efficient allocations
    • Job scheduling for energy efficient systems
Barcelona Supercomputing Center - Centro Nacional de Supercomputación

Source URL (retrieved on 23 Mar 2023 - 21:17): https://www.bsc.es/ca/research-development/research-areas/programming-models/dynamic-resource-management-hpc