Skip to Main content Skip to Navigation
Reports

Topology-aware resource management for HPC applications

Abstract : The Resource and Job Management System (RJMS) is a crucial system software part of the HPC stack. It is responsible for efficiently delivering computing power to applications in supercomputing environments. Its main intelligence relies on resource selection techniques to find the most adapted resources to schedule the users' jobs. Improper resource selection operations may lead to poor performance executions and global system utilization along with increase of system fragmentation and jobs starvation. These phenomenas play a role in the increase of the platforms' total cost of ownership and should be minimized. This paper introduces a new topology-aware re- source selection algorithm to determine the best choice among the available nodes of the platform based upon their position within the network and taking into account the applications commu- nication matrix. To validate our approach, we integrated this algorithm as a plugin for Slurm, a popular and widespread HPC resource and job management system (RJMS). We validated our plugin with different optimization schemes by comparing with the default Slurm algorithm using both emulation of a large-scale platform, and by carrying out experiments in a real cluster.
Complete list of metadatas

Cited literature [22 references]  Display  Hide  Download

https://hal.inria.fr/hal-01275270
Contributor : Guillaume Mercier <>
Submitted on : Monday, April 4, 2016 - 3:31:46 PM
Last modification on : Monday, May 4, 2020 - 11:38:48 AM
Document(s) archivé(s) le : Monday, November 14, 2016 - 3:21:32 PM

File

RR-8859.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01275270, version 2

Citation

Yiannis Georgiou, Emmanuel Jeannot, Guillaume Mercier, Adèle Villiermet. Topology-aware resource management for HPC applications. [Research Report] RR-8859, Inria Bordeaux Sud-Ouest ; Bordeaux INP; LaBRI - Laboratoire Bordelais de Recherche en Informatique. 2016, pp.17. ⟨hal-01275270v2⟩

Share

Metrics

Record views

679

Files downloads

1639