Model-driven Con fi guration of Cloud Computing Auto-scaling Infrastructure

Cloud computing uses virtualized computational resources to allow an application’s computational resources to be provisioned on-demand. Autoscaling is an important cloud computing technique that dynamically allocates computational resources to applications to precisely match their current loads. This paper presents a model-driven engineering approach to optimizing the configuration and cost of cloud auto-scaling infrastructure. The paper provides the following contributions to the study of model-driven configuration of cloud autoscaling infrastructure: (1) it shows how virtual machine configurations can be captured in feature models, (2) it describes how these models can be transformed into constraint satisfaction problems (CSPs) for configuration and cost optimization, (3) it shows how optimal auto-scaling configurations can be derived from these CSPs with a constraint solver, and (4) it presents a case-study showing the cost reduction produced by this model-driven approach.