ICE-T: RC: Multi-Domain Multi-Broker Elastic Optical Networks with Cognitive Functionalities

University Of California-Davis, Davis CA

Investigators

Abstract

The proposed project seeks an innovative approach to the design, control and management of next generation Internet core networks. A new inter-domain networking paradigm is considered, where broker agents compete freely with each other to provide attractive inter-networking services to Autonomous Systems (ASes), while ASes also choose broker services suitable for their inter-networking needs. The project develops a market-driven multi-broker-assisted multi-domain framework that can coordinate -- in a distributed, incentivized manner - while providing complete autonomy and freedom to both brokers and ASes. Through market-driven incentives and competitions, the brokers and the ASes will pursue mutually beneficial services, which will in turn continuously spur innovative solutions and services. The new inter-domain networking architecture with multiple brokers will facilitate automated, Quality-of-Service aware, and impairment-responsive services. The project leverages cognitive functionalities based on machine learning to optimize end-to-end service provisioning. The project seeks to architect, simulate, prototype, and experiment to improve system (a) scalability, (b) end-to-end performance, (c) robustness and adaptability, and (d) manageability and coordination. The project will leverage novel and emerging cognitive techniques for dynamic traffic prediction and adaptive provisioning, quality of transmission estimation and fault detection across multiple domains. This project extends a research collaboration between the University of California, David, the Universitat Politecnica de Cataluna (UPC), Spain, and Consorzio Nazionale and Interuniversitario per le Telecomunicazioni (CNIT), Italy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.