Horizontal Scaling in Cloud Using Contextual Bandits

One characteristic of the Cloud is elasticity: it provides the ability to adapt resources allocated to applications as needed at run-time. 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.