Adaptive, Model-driven Autoscaling for Cloud Applications

Applications with a dynamic workload demand need access to a flexible infrastructure to meet performance guarantees and minimize resource costs. While cloud computing provides the elasticity to scale the infrastruc-ture on demand, cloud service providers lack control and visibility of user space applications, making it difficult to accurately scale the underlying infrastructure. Thus, the burden of scaling falls on the user. In this paper, we propose a new cloud service, Depend-able Compute Cloud (DC2), that automatically scales the infrastructure to meet the user-specified performance re-quirements. DC2 employs Kalman filtering to automati-cally learn the (possibly changing) system parameters for each application, allowing it to proactively scale the in-frastructure to meet performance guarantees. DC2 is de-signed for the cloud- it is application-agnostic and does not require any offline application profiling or bench-marking. Our implementation results on OpenStack us-ing a multi-tier application under a range of workload traces demonstrate the robustness and superiority of DC2 over existing rule-based approaches.