Market-based hierarchical resource management using machine learning
DSOM'07 Proceedings of the Distributed systems: operations and management 18th IFIP/IEEE international conference on Managing virtualization of networks and services
Towards microeconomic resources allocation in overlay networks
AMT'10 Proceedings of the 6th international conference on Active media technology
A non-strategic microeconomic model for single-service multi-rate application layer multicast
ICICA'10 Proceedings of the First international conference on Information computing and applications
An economic case for end system multicast
FIS'10 Proceedings of the Third future internet conference on Future internet
Journal of Network and Systems Management
Journal of Parallel and Distributed Computing
Toward microeconomic allocation of resources in multi-service overlay networks
Journal of Computer and Systems Sciences International
Rate allocation in overlay networks based on theory of firm consumer
HPCA'09 Proceedings of the Second international conference on High Performance Computing and Applications
| Hi-index | 0.07 |
Rather than managing their heterogeneity and dynamic behavior through centralized intervention, overlay nodes can be programmed to self-organize and self-manage the network. To achieve the highest performance within a service overlay, they are further expected to self-optimize the network, by cooperatively providing and allocating resources in an optimal manner. However, since nodes are inherently selfish about resources they contribute or consume, self-optimization could not be achieved if they are not given the correct incentives. In this paper, we investigate the effectiveness of a market-based incentive mechanism in directing nodes' behavior and enabling self-optimizations. We have designed an intelligent market model for a service overlay network, based on which individual nodes, being service producers and consumers, determine their own resource contributions, consumptions, or service prices based on their own utility maximization goals. We also propose optimal decision making solutions for nodes to achieve their self-interests; in particular, service providers are provided with a control-based pricing solution based on system identification techniques. With the multicast streaming application as an example, we show through simulations that, even when selfish nodes all seek their maximal utilities, the resulting network still achieves close-to-optimal performance in both steady and dynamic states. The results also indicate that, by encouraging nodes to behave selfishly and intelligently in a designed market, self-optimization in other autonomic systems may be facilitated in the presence of node selfishness.