Skip to Main content Skip to Navigation
Conference papers

Horizontal Scaling in Cloud Using Contextual Bandits

Abstract : One characteristic of the Cloud is elasticity: it provides the ability to adapt resources allocated to applications as needed at runtime. This capacity relies on scaling and scheduling. In this article online horizontal scaling is studied. The aim is to determine dynamically applications deployment parameters and to adjust them in order to respect a Quality of Service level without any human parameters tuning. This work focuses on CaaS (container-based) environments and proposes an algorithm based on contextual bandits (HSLinUCB). Our proposal has been evaluated on a simulated platform and on a real Kubernetes’s platform. The comparison has been done with several baselines: threshold based auto-scaler, Q-Learning, and Deep Q-Learning. The results show that HSLinUCB gives very good results compared to other baselines, even when used without any training period.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03407934
Contributor : Patricia Stolf Connect in order to contact the contributor
Submitted on : Thursday, October 28, 2021 - 5:28:23 PM
Last modification on : Thursday, November 4, 2021 - 4:05:34 AM

Identifiers

Citation

David Delande, Patricia Stolf, Raphaël Feraud, Jean-Marc Pierson, André Bottaro. Horizontal Scaling in Cloud Using Contextual Bandits. 27th International Conference on Parallel and Distributed Computing (Euro-Par 2021), INESC-ID, Lisbon, Portugal; Instituto Superior Técnico (IST),Lisbon, Portugal, Sep 2021, Lisbon (Online Event), Portugal. pp.285-300, ⟨10.1007/978-3-030-85665-6_18⟩. ⟨hal-03407934⟩

Share

Metrics

Les métriques sont temporairement indisponibles