Virtual machine experience design : a predictive resource allocation approach for cloud infrastructures

Abstract : One of the main challenges for cloud computing providers remains to offer trustable performance for all users, while maintaining an efficient use of hardware and energy resources. In the context of this CIFRE thesis lead with Outscale, apublic cloud provider, we perform an in-depth study aimed at making management algorithms use new sources of information. We characterize Outscale’s workload to understand the resulting stress for the orchestrator, and the contention for hardware resources. We propose models to predict the runtime of VMs based on features which are available when they start. We evaluate the sensitivity with respect to prediction error of a VM placement algorithm from the literature that requires such predictions. We do not find any advantage in coupling our prediction model and the selected algorithm, but we propose alternative ways to use predictions to optimize the placement of VMs.
Complete list of metadatas

Cited literature [174 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-02372649
Contributor : Abes Star <>
Submitted on : Wednesday, November 20, 2019 - 3:12:28 PM
Last modification on : Saturday, December 21, 2019 - 5:12:40 AM

File

ThesePERENNOU.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-02372649, version 1

Collections

Citation

Loïc Pérennou. Virtual machine experience design : a predictive resource allocation approach for cloud infrastructures. Distributed, Parallel, and Cluster Computing [cs.DC]. Conservatoire national des arts et metiers - CNAM, 2019. English. ⟨NNT : 2019CNAM1246⟩. ⟨tel-02372649⟩

Share

Metrics

Record views

329

Files downloads

162