Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Client-side scheduling based on application characterization on Kubernetes

Medel, Víctor, Tolón, Carlos, Arronategui, Unai, Tolosana-Calasanz, Rafael, Bañares, José Ángel and Rana, Omer Farooq ORCID: https://orcid.org/0000-0003-3597-2646 2017. Client-side scheduling based on application characterization on Kubernetes. Presented at: GECON 2017: International Conference on the Economics of Grids, Clouds, Systems, and Services, Biarritz, France, 19-21 September 2017. Published in: Pham, C., Altmann, J. and Bañares, J. eds. Economics of Grids, Clouds, Systems, and Services.GECON 2017. Lecture Notes in Computer Science. Lecture Notes in Computer Science , vol.10537 Springer, pp. 162-176. 10.1007/978-3-319-68066-8_13

Full text not available from this repository.

Abstract

In container management systems, such as Kubernetes, the scheduler has to place containers in physical machines and it should be aware of the degradation in performance caused by placing together containers that are barely isolated. We propose that clients provide a characterization of their applications to allow a scheduler to evaluate what is the best confguration to deal with the workload at a given moment. The default Kubernetes Scheduler only takes into account the sum of requested resources in each machine, which is insufficient to deal with the performance degradation. In this paper, we show how specifying resource limits is not enough to avoid resource contention, and we propose the architecture of a scheduler, based on the client application characterization, to avoid the resource contention.

Item Type: Conference or Workshop Item (Paper)
Date Type: Publication
Status: Published
Schools: Computer Science & Informatics
Publisher: Springer
ISBN: 9783319680651
ISSN: 0302-9743
Date of First Compliant Deposit: 12 October 2017
Date of Acceptance: 1 August 2017
Last Modified: 03 Nov 2022 09:37
URI: https://orca.cardiff.ac.uk/id/eprint/105487

Citation Data

Cited 24 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item Edit Item