A Market-based Framework for Resource Management in Cloud Federation

• Published in: Journal of grid computing 2023-03, Vol.21 (1), p.3, Article 3
• Main Authors: Ramezani, Faeze, Abrishami, Saeid, Feizi, Mehdi
• Format: Article
• Language: English
• Subjects: Algorithms; Cloud computing; Communication; Computer Science; Cooperation; Design; Management of Computing and Information Systems; Processor Architectures; Provisioning; Resource management; User Interfaces and Human Computer Interaction; Workloads
• ISSN: 1570-7873; 1572-9184
• DOI: 10.1007/s10723-022-09635-w

A federated cloud is a form of the inter-cloud environment in which cloud service providers cooperate for better scalability and service provisioning. By communicating with other cloud providers and sharing resources, federated members gain more advantages from utilizing a comprehensive resource management system. There are different setups for federated clouds, in terms of federation formation and member interactions. Consequently, difficulties arise since resource management must be dealt with as a general entity. The current paper introduces a generic resource management framework for inter-federation resource management that can be used for a wide range of cloud federation models. By considering the federated cloud as a market for trading resources, the present study proposes a market-based framework to manage the various types of federated clouds. The main components of the proposed framework are based on market management models. This generic framework is able to cover a variety of centralized cloud federation models that divide the federated cloud life cycle into time slots. Finally, the present work introduces a resource management model which is compatible with the proposed framework. The model is implemented with Java and NetBeans IDE 8.2 and evaluated using the FederatedCloudSim 2.0 toolkit. The results are discussed in order to evaluate the proposed framework and the selected market algorithms.