Cloud computing has gained wide popularity in the past few years as it delivers computing services on-demand on a pay-as-you-go basis. The rapid expansion in scale and complexity of cloud computing resources and their services have made monitoring, controlling, and provisioning cloud computing resources a challenging and complex task. As there is no first-principles' dynamic model for cloud computing systems, the grey-box identification technique offers a possible alternative. To control and provision the cloud computing resources, tools from control-theoretic techniques provide a rigorous mathematical foundation and systematic mechanisms.

In this talk, first, we will discuss the system identification of the cloud computing system via the linear-parameter-varying (LPV) approach. Secondly, a decentralized LPV-MPC controller with the workload prediction will be discussed, for ensuring the performance under the time-varying workload. Then, a heuristic-based approach will be discussed for load balancing across different servers. Finally, the result of the proposed approach on a real-time private cloud setup will be presented.