A Novel Truthful and Fair Resource Bidding Mechanism for Cloud Manufacturing

摘要

Cloud manufacturing is an emerging service-oriented and market-oriented networked manufacturing paradigm. Resource scheduling of cloud manufacturing is different from those of other manufacturing paradigms due to some distinctive characteristics, such as high concurrency of multi-service and multi-task, high heterogeneity and high dynamics of resources, incomplete information of cloud services and user-centric services. These characteristics make the resource scheduling problem of cloud manufacturing more complicated. Efficient resource scheduling and economic benefit are the major concerns of users in cloud manufacturing environment. A novel market-based continuous bidding mechanism is proposed in this paper by applying game theory. In this mechanism, the overall benefit of both the cloud service provider and the cloud service demander is considered as the optimization target. The constraints are the task delivery time and budget. For decomposable tasks, the constraints for each subtask are also considered in resource bidding. Service demanders can bid continuously until the corresponding service is obtained or the bidding constraint is violated. Experiments demonstrate the proposed resource bidding mechanism is efficient in cloud service scheduling and can ensure the cloud manufacturing market is truthful and fair.