Hierarchical and Frequency-Aware Model Predictive Control for Bare-Metal Cloud Applications

Bare-metal cloud provides a dedicated set of physical machines (PMs) and enables both PMs and virtual machines (VMs) on the PMs to be scaled in/out dynamically. However, to increase efficiency of the resources and reduce violations of service level agreements (SLAs), resources need to be scaled quickly to adapt to workload changes, which results in high reconfiguration overhead, especially for the PMs. This paper proposes a hierarchical and frequency-aware auto-scaling based on Model Predictive Control, which enable us to achieve an optimal balance between resource efficiency and overhead. Moreover, when performing high-frequency resource control, the proposed technique improves the timing of reconfigurations for the PMs without increasing the number of them, while it increases the reallocations for the VMs to adjust the redundant capacity among the applications; this process improves the resource efficiency. Through trace-based numerical simulations, we demonstrate that when the control frequency is increased to 16 times per hour, the VM insufficiency causing SLA violations is reduced to a minimum of 0.1% per application without increasing the VM pool capacity.

I. Introduction

Bare-metal cloud offers infrastructure as a service (IaaS) in which a customer uses a dedicated set of physical servers (also called physical machines (PMs)) on a pay-per-use basis [1]. Existing on-premises types of deployment for business-critical applications, such as web-based applications like e-mail and collaboration [2], often use dedicated PM clusters to handle peak workload to avoid violating service level agreements (SLAs) and to satisfy manageability of software licenses and requirements for audit of compliance and security, which can cause the applications to become over-provisioned and underutilized most of the time [3]. We suppose that an application provider rents such a PM cluster from a bare-metal cloud provider to improve resource efficiency, creates a VM pool on the cluster, and hosts business-critical applications on the pool without changing existing management policies. In this paper, we try to develop an optimal resource allocation mechanism for such applications in bear-metal cloud environments.