Title: A fairness-enhanced resource trading system in federated cloud environments

Authors: Tienan Zhang

Addresses: School of Computer and Communication, Hunan Institute of Engineering, Hunan Province, China

Abstract: In federated cloud system, resource provisioning and allocation across multiple providers become a challenging issue. However, most of existing studies focused on how to improve resource utility or resource revenue, while ignores the efficiency and fairness of the resource market. In this study, we first introduce the definitions of efficient market and fairness criterion based on the classical economic theory; then, we design a resource trading protocol which is capable of offering topology-wide efficiency and fairness under certain payment rules in resource market. Theoretical analysis indicates that the proposed resource trading model is beneficial for pushing resources to resource providers who value them more, which means that resource providers can increase social welfare by repeatedly performing ration deals. To evaluate the effectiveness of the proposed trading model, experiments are conducted in a real-world cloud platform. The results show that it can significantly improve the resource revenues for providers and make better task mapping decisions.

Keywords: cloud computing; resource trading model; multi-agent model; virtual machine.

DOI: 10.1504/IJNVO.2020.105522

International Journal of Networking and Virtual Organisations, 2020 Vol.22 No.2, pp.183 - 198

Received: 28 Apr 2018
Accepted: 12 Sep 2018

Published online: 03 Mar 2020 *

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