Authors: Tienan Zhang
Addresses: School of Computer and Communication, Hunan Institute of Engineering, Hunan Province, China
Abstract: In cloud computing, quality of service (QoS) has become an important issue either from the perspective of cloud providers or users. In this paper, we present a lightweight framework, namely cloud QoS ranking and controlling (CQRC), which integrates a QoS ranking service and a fine-grained QoS controller. The QoS ranking service provides an effective approach to quickly evaluating the QoS level of any cloud resources or services based on historical performances; the QoS controller is designed by using feedback control technique and is capable of self-tuning the underlying virtual machine (VM) configurations. The proposed framework is implemented and evaluated in a real-world cloud platform through a set of experiments and the results indicate that its QoS ranking service can significantly reduce the frequency of QoS violations and the QoS controller can achieve better tradeoffs between performance and cost for cloud providers.
Keywords: resource virtualisation; virtual machine; scheduling algorithm; cloud computing.
International Journal of Networking and Virtual Organisations, 2019 Vol.21 No.4, pp.455 - 470
Received: 23 Sep 2017
Accepted: 12 Jan 2018
Published online: 15 Nov 2019 *