The first part of the paper will look into upgrading the operating system of a cluster from CentOS 7 to CentOS 8 in a bid to test the stability and production readiness of the CentOS 8.0. The second part of the paper focuses on transiting from Torque and Maui to Slurm (batch system) porting the same conguration to achieve the same or improved scheduling from Slurm with shorter waiting time for jobs as well as maintaining the same load or obtaining higher capacity on the cluster.
Modernizing a high performance cluster: a case for ICMS at Temple University(2020 Feb 14).
|Autori:||Nguruwe, Herbert Kudakwashe|
|Titolo:||Modernizing a high performance cluster: a case for ICMS at Temple University|
|Relatore/i esterni:||Posada, Fernando; Kohlmeyer, Axel|
|Data di pubblicazione:||14-feb-2020|
|Aree SISSA:||Laboratorio Interdisciplinare|
|Appare nelle tipologie:||8.4 Master thesis in High Performance Computing (HPC)|