This project is for a workshop on "Scaling Terabit Networks: Breaking Through Capacity Barriers and Lowering Cost with New Architectures and Technologies," to be held at the Optical Society of America (OSA) headquarters in Washington, DC, on September 19-20, 2013. The goal of the workshop is to develop a set of "grand challenges" for the optical networking community. This will be accomplished by: exploring how new network architectures driven by changing traffic patterns, virtualization and programmability drive new requirements for the optics, and how emerging photonic components can help scale these architectures to terabit capacities; rethinking the operating system interface and application interface to the network; exploring what physical layer phenomena and characteristics need to be communicated to the higher layers and what higher-layer functions can benefit from physical layer intelligence; and understanding how optics and electronics will be used together in the future and what their respective metrics will be when used together in an optical networking context.
In addressing these technical areas, the workshop will be organized to achieve the following objectives: identify the technological challenges facing terabit networks and continued network scaling; establish key metrics, targets, and capabilities for efficient and scalable terabit networking; set research priorities and requirements within optical networking to enable efficient and scalable terabit networking; and identify application drivers and early adopters of this new technology.
Ultra-high capacity optical networks are the key enabler of the ongoing revolution in data-driven science. They are also part of critical infrastructure technologies that the United States needs to maintain national competitiveness and national security. By informing both the U.S. research community as well as NSF and other Federal agencies about key research problems and opportunities, this workshop will support U.S. national interests. The final workshop report will be made publicly available.
To address the challenge of scaling terabit networks, the Center for Integrated Access Networks, the Optical Society, and the National Science Foundation convened a workshop in September, 2013. Participants in the workshop were selected from academia, industry, and governmental organizations, and represent the entire network stack. The technical program of the workshop was organized into a series of sessions that built toward the development of a prioritized list of research challenges specifically identifying candidate topics to form ‘grand research challenges’. Three plenary speakers gave their perspectives on key issues in technical areas related to scaling future networks. Underpinning the discussion of future networks is the identification of several concrete trends, obstacles, and opportunities. The workshop participants identified the following key network evolution issues: (1) Optical technology is penetrating further to the edge - as network hosts scale from 10 Gbps to 100 Gbps and beyond, optics is already reaching the top-of-rack switch in datacenters and migrating further to end user and home networks. Advanced optical network technologies are needed for a wide range of new application spaces including data center, metro, aggregation, mobile backhaul, and access networks, each of which has unique requirements and design constraints that are often very different from those of traditional backbone networks upon which most fiber systems are based today. (2) New applications and services are demanding new architectures, specifically to address "Big Data" and large scientific deployments. The consolidation of computing into high-density data centers will result in dramatic changes to access and aggregation networks. (3) The rise of data centers will require new network topologies as well as novel distributed control paradigms. In this context, software-defined networking is creating a critical need for (4) new end-to-end control planes and virtualization. These control planes will present compute, storage, and networking as a seamless, reconfigurable resource, rather than as discrete components. (5) The proliferation of mobile networks, are driving increased capacity requirements with a greater need for optics, leading potentially to new coding, advanced modulation formats, high capacity small cells, and distributed and smart antenna techniques. (6) Spectral efficiency is reaching the Shannon limits of optical transmission, and is leading to (7) increased parallelism in optical networks, which is fueling the transition to photonic integration and scalable platforms such as CMOS compatible Silicon photonics. New integrated and parallel photonics, tightly embedded with the electronics, may enable greater computer system disaggregation. Thus, not only are optical networks being used in new ways and for new applications, but the optical technologies themselves are undergoing an evolution, opening the door for innovation and potentially accelerating the impact of research. The workshop participants identified the key research priorities grouped around three grand challenges and the need for a national scale multi-user networking test-bed: 1. Programmable, Virtualized and Intelligent Optical Networks for the Future Internet 2. Cross-Layer Optical Network Architectures for Datacenters & Cloud Computing 3. Clean Slate Architectures and Component Technologies for Optical Networks The workshop findings were detailed in a final report that is available on the workshop website: www.osa.org/en-us/meetings/osa_incubator_meetings/scaling_terabit_networks_breaking_through_capacity/ Workshop outcomes were also featured in the September 2014, Optics and Photonics News article entitled: 'Optical Networks Come of Age'.
