The purpose of the project is to create dynamic, deterministic, and manageable end-to-end network transport services for high-end e-Science applications. This service capability will build on the success and reach of the IP infrastructure, and will complement it with deterministic end-to-end multi-protocol services spanning multiple administrative domains and a variety of conventional campus network technologies. The project will create a Generalized MultiProtocol Label Switching (GMPLS) capable optical core network. Optical transport and switching equipment acting as Label Switching Routers (LSRs) will provide deterministic network resources at the packet, wavelength, and fiber cross connect levels. The all-optical capabilities proposed for this core network will be based on connection and resource management mechanisms defined in GMPLS and will model many of the functions of an inter-regional, national, or even global wavelength based Research and Education (R&E) networks. Attached to the core infrastructure will be experimental campus networks carved out from collaborating institutional campus infrastructure. Network requirements will be driven by tight integration with application team members.
An important part of the project will be the development of the necessary software components to address unresolved GMPLS-related issues as well as a host of practical end-to-end issues. The architecture describes a Network Aware Resource Broker (NARB) that will peer with Interior Gateway Protocol/Exterior Gateway Protocol (IGP/EGP) routing protocols to advertise inter-domain service capabilities, perform inter-domain path computation, resource allocation, scheduling, provisioning, proxy signaling services, and authorization/accounting for the requested network services. The architecture of the NARB function will integrate emerging advanced services from the NSF Middleware Initiative (NMI) program and Grid services, particularly for security and advanced scheduling. The project includes collaboration with specific e-Science applications support facilities that will be adapted to and directly benefit from the experimental infrastructure. The electronic-Very Long Baseline Interferometry (e-VLBI) project from MIT Haystack Observatory will utilize the experimental network to acquire and correlate radio-telescope data in real time as a pilot to demonstrate effective coordination of multiple remote instruments. USC/ISI East, University of Maryland Visualization and Presentation Lab, and NASA Goddard Space Flight Center will collaborate to use the resources as part of ongoing efforts to integrate streaming HDTV, Access Grid Nodes, and interactive remote 3D visualizations.
The intellectual merit of this project includes greatly expanding the knowledge and technologies relating to intelligent management of dynamic optical transport networks. This will be necessary in order for the larger R&E community to realize maximum benefits from its current efforts to procure a national footprint of lambda and fiber resources. Broader impacts will be realized via the participation of regional universities that will directly enhance the research and education infrastructure and provide educational opportunities for undergraduate and graduate students as well as involve women and underrepresented minorities as interns in the activities.
Webmaster Shujia Gong
Please send your comments to sgong@gmu.edu