Book Description

Optical transport networks based on wavelength division multiplexing (WDM) are considered to be the most appropriate choice for future Internet backbone. On the other hand, future DOE networks are expected to have the ability to dynamically provision on-demand survivable services to suit the needs of various high performance scientific applications and remote collaboration. Since a failure in aWDMnetwork such as a cable cut may result in a tremendous amount of data loss, efficient protection of data transport in WDM networks is therefore essential. As the backbone network is moving towards GMPLS/WDM optical networks, the unique requirement to support DOE's science mission results in challenging issues that are not directly addressed by existing networking techniques and methodologies. The objectives of this project were to develop cost effective protection and restoration mechanisms based on dedicated path, shared path, preconfigured cycle (p-cycle), and so on, to deal with single failure, dual failure, and shared risk link group (SRLG) failure, under different traffic and resource requirement models; to devise efficient service provisioning algorithms that deal with application specific network resource requirements for both unicast and multicast; to study various aspects of traffic grooming in WDM ring and mesh networks to derive cost effective solutions while meeting application resource and QoS requirements; to design various diverse routing and multi-constrained routing algorithms, considering different traffic models and failure models, for protection and restoration, as well as for service provisioning; to propose and study new optical burst switched architectures and mechanisms for effectively supporting dynamic services; and to integrate research with graduate and undergraduate education. All objectives have been successfully met. This report summarizes the major accomplishments of this project. The impact of the project manifests in many aspects: First, the project addressed many essential problems that arisen in current and future WDM optical networks, and provided a host of innovative solutions though there was no invention or patent filing. This project resulted in more than 2 dozens publications in major journals and conferences (including papers in IEEE Transactions and journals, as well as a book chapter). Our publications have been cited by many peer researchers. In particular, one of our conference papers was nominated for the best paper award of IEEE/Create-Net Broadnets (International Conference on Broadband Communications, Networks, and Systems) 2006. Second, the results and solutions of this project were well received by DOE Labs where presentations were given by the PI. We hope to continue the collaboration with DOE Labs in the future. Third, the project was the first to propose and extensively study multicast traffic grooming, new traffic models such as sliding scheduled traffic model and scheduled traffic model. Our research has sparkled a flurry of recent studies and publications by the research community in these areas. Fourth, the project has benefited a diverse population of students by motivating, engaging, enhancing their learning and skills. The project has been conducted in a manner conducive to the training of students both at graduate and undergraduate levels. As a result, one Ph. D., Dr. Abdur Billah, was graduated. Another Ph. D. student, Tianjian Li, will graduate in January 2007. In addition, four MS students were graduated. One undergraduate student, Jeffrey Alan Shininger, completed his university honors project. Fifth, thanks to the support of this ECPI project, the PI has obtained additional funding from the National Science Foundation, the Air Force Research Lab, and other sources. A few other proposals are pending. Finally, this project has also significantly impacted the curricula and resulted in the enhancement of courses at the graduate and undergraduate levels, therefore strengthening the bond between research and education.