Title: A new scheme for improving the utilisation of nested-cloud resources

Authors: Yuan Ni; Zhiwei Zhang; Xiaoling Tao

Addresses: State Key Laboratory of Integrated Service Networks (ISN), Xidian University, Xian, 710071, China ' State Key Laboratory of Integrated Service Networks (ISN), Xidian University, Xian, 710071, China ' Guangxi Cooperative Innovation Center of Cloud Computing and Big Data, Guangxi Colleges and Universities Key Laboratory of Cloud Computing and Complex Systems, Guilin University of Electronic Technology, Guilin, 541004, China

Abstract: In recent years, cloud computing has played an important role in IT fields. It brings benefits such as high computing efficiency, cost saving, provisioning on-demand of computing resources and high utilisation of the hardware. However, with more fields and businesses involved in cloud computing, the inherent problems in its structure are restricting the further development of cloud computing. And the most serious problem is how to maximumly achieve the high-usage of the cloud resources. In this paper, we present a scheme based on multi-layer nested virtualisation technology. It can make full use of the allocations of a virtual machine (VM) provided by an infrastructure-as-a-service (IaaS) provider. Furthermore, we also propose a task scheduling approach for nested cloud environment correspondingly. The task scheduling algorithm considers both the computing efficiency and the monetary costs paid for the cloud service. In order to justify our proposal, we use KVM as unmodified multiple hypervisor to run multi-layer VMs on x86 platform. We also do simulations to test our task scheduling algorithm. And the result shows encouraging support for our proposal.

Keywords: cloud architecture; multi-layer nested virtualisation; task scheduling; KVM; Intel-VT.

DOI: 10.1504/IJES.2019.098291

International Journal of Embedded Systems, 2019 Vol.11 No.2, pp.170 - 181

Received: 14 Oct 2016
Accepted: 06 Feb 2017
Published online: 13 Mar 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article