Cluster Workload Analytics Revisited - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Cluster Workload Analytics Revisited

Résumé

Community compute clusters are now common in big organizations such as in universities and industries. These are excellent targets for analyzing resource usage and failure events because of the diverse user base and corresponding diversity of workloads. We introduce an public repository that hosts such rich dataset and would be useful for systems and dependability research communities. In this paper, we discuss our goals, provide details of the dataset, and present some statistical analysis.
Fichier principal
Vignette du fichier
Cluster Workload Analytics Revisited.pdf (99.04 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01316530 , version 1 (17-05-2016)

Identifiants

  • HAL Id : hal-01316530 , version 1

Citer

Subrata Mitra, Suhas Javagal, Todd Gamblin, Adam Moody, Stephen Harrell, et al.. Cluster Workload Analytics Revisited. Fast Abstract in the 46th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, Jun 2016, Toulouse, France. ⟨hal-01316530⟩
111 Consultations
135 Téléchargements

Partager

Gmail Facebook X LinkedIn More