This project will create and demonstrate experimentally key features of a new optical network architecture, 'HyperFlow' which is a hybrid future Internet architecture that include key designs for both hardware and algorithms. The network is designed to be dynamic (both agile and adaptive), and significantly more cost effective and power efficient for future growth in data volumes and number of users. The architecture relies on a novel optical network infrastructure comprising new transport mechanisms and a new comprehensive control mechanism including the physical hardware, algorithms and applications.

Between locations that exchange large volumes of data - say, Los Angeles and New York City ? the flow switching mechanism of HyperFlow would establish a dedicated path across the network. The allotment of bandwidth would change constantly. As traffic between New York and Los Angeles increased, new, dedicated wavelengths would be recruited to handle it; as the traffic tailed off, the wavelengths would be relinquished. The goal is to develop network management protocols that can perform new session allocations in a matter of sub-seconds. HyperFlow can easily increase the data rates of optical networks 100-fold with the new network management scheme to be addressed in this program, with corresponding decrease in cost per bit and power consumption as well.

Intellectual Merit: HyperFlow creates a complete hybrid flow/IP network architecture spanning local to wide areas, sharing the same unified control plane. The network provides both guaranteed end-to-end multi-Gbps flow service and conventional Internet Packet services throughout the network down to individual users. To achieve that goal, HyperFlow includes novel access and long haul core network technologies designed to support end-to-end flows as well as conventional IP network services. This approach achieves a breakthrough with respect to previous approaches by providing both services end-to-end with a unified control plane. HyperFlow includes major innovations based on highly efficient, agile and control of pools of shared tunable lasers and passive optical devices, a fast MAC, partitioned routing strategy and a new transport protocol for flows.

Broader impact: If successful the proposed HyperFlow technology will have a profound influence on the way the future Internet is designed and perceived by the society, as the new on-demand cost-effective gigabit service will enable applications that are hard or costly to support today with existing broadband access, such as instantly-available 3D distributed virtual-reality systems, cloud computing and instant file transfer services including telemedicine, education, 3D movies, concerts, sporting events, and government services. The concept of the new Hyperflow development will be brought into industry forums such as IETF/IRTF during the project, so that the best minds of the industry can be leveraged for conceptual validation.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Network Systems (CNS)
Application #
1111374
Program Officer
john brassil
Project Start
Project End
Budget Start
2011-08-01
Budget End
2016-03-31
Support Year
Fiscal Year
2011
Total Cost
$830,922
Indirect Cost
Name
Stanford University
Department
Type
DUNS #
City
Stanford
State
CA
Country
United States
Zip Code
94305