CISE ENCOURAGES COLLABORATIONS TO TRANSFORM NETWORK SCIENCE AND ENGINEERING

BACKGROUND AND MOTIVATION

Fueled by continuing advances in computing, information and communication technologies, the Internet has become integral to our lives and vital to the operation of important sectors of our society. It has evolved from a small-scale network of computers to a large-scale, highly-engineered network of networks operating over a complex communications infrastructure of heterogeneous devices and physical components. Because the fragmented nature of its underlying infrastructure and the intricate layers of its protocols are largely invisible to users, new classes of participatory networks in which users actively and spontaneously communicate, create content, and share knowledge in a variety of ways have emerged.

The integration of information networks and social applications underscore the need for new insights into both the technical capabilities of the information networks and the ways in which people engage or are affected by the use of the information and content these networks provide. As critical as these information networks and the distributed applications they enable are to our lives and diverse sectors of our society, we have little rigorous knowledge for understanding their structure, dynamics and holistic behaviors. This lack of a formal basis underscores the need for a "network science and engineering" that articulates the principles and methodologies necessary to gain better understanding of how to effectively design robust and secure large-scale networks in ways that allow for sustainable performance and predictable behaviors. Such an endeavor is by its very nature interdisciplinary.

The purpose of this announcement is to alert the Computer and Information Science and Engineering (CISE) research and education communities to current funding opportunities and future directions intended to enable and sustain scientific collaborations that promise to transform network science and engineering.

OBJECTIVES

To encourage researchers to develop fundamentally new ways of thinking about the structure, dynamics and behaviors of large-scale networks and the critical services and applications they support, CISE is developing a comprehensive strategy to enable activities that encourage inter-disciplinary collaborations among researchers in multiple areas of computer and information science, as well as with researchers in other relevant fields such as engineering, mathematics, economics, and the social sciences. Considerable knowledge related to different aspects of information and communications networks has been generated by each of these communities. The objective is to synthesize these bodies of knowledge into new fundamental knowledge required for the design, construction and operation of robust and sustainable global networks on which our safety and security can reliably depend.

As a first step, each division in CISE has encouraged increased communication and collaboration in all areas of network science and engineering research. To summarize, in Fiscal Year 2007, the Future Internet Design (FIND) Program in the Division of Computer and Network Systems (CNS) solicited projects on clean-slate approaches to network architecture and encouraged projects inspired by economics, biology, information privacy, cyber trust, and other areas of science beyond networking and distributed systems. The Scientific Foundations for Internet's Next Generation (SING) Program in the Division of Computer and Communications Foundations (CCF) encouraged theory of communications and signal processing researchers to consider new paradigms and foundational research on network science. In Fiscal Year 2008, the Next Generation Networked Information (NGNI) Program is encouraging Division of Information and Intelligent Systems (IIS) researchers to play a role in future network technology by envisioning new classes of information systems that would take advantage of future networking environments. Interdisciplinary proposals submitted to these programs will receive appropriate review in relevant panels.

Below, we discuss three current funding opportunities designed to enable collaborations across network science and engineering. Pointers are provided to more information about these opportunities on the CISE web pages. At the end, we discuss how these programs will bridge to the future.

Future Internet Design (FIND)

FIND is a major new long-term initiative of the Networking Technology and Systems (NeTS) Program. The intellectual scope of the FIND program is wide. FIND research addresses questions such as:

• How can we design a network that is fundamentally more secure and available than today's Internet? How would we conceive the security problem if we could start from scratch?
• How might such functions as information dissemination, location management or identity management best fit into new network architecture?
• What will be the long-term impact of new technologies such as advanced wireless and optics?
• How will economics and technology interact to shape the overall design of a future network?
• How do we design a network that preserves a free and open society?

The FIND Program is designed as a multi-year, three-phase program based on a new conceptualization of networking research that is unconstrained by the properties of today's Internet. The goal of Phase I is to identify the functional elements of the Future Internet and their interactions. Investigators have been encouraged to propose and evaluate requirements for new architectures, to reason carefully about architectural responses to these requirements, and to propose, develop and demonstrate ideas that might be a part of new architectures. FIND Phase I is supporting research across a broad range of topics including, but not limited to:

• Create new core functionality and higher-level service architectures;
• Design for architectural implications of new optical, wireless, sensor network and embedded computing technologies;
• Design for holistic networks; and
• Develop theories of network architecture security, manageability, social requirements, and real-time, uninterruptible services.

In Phase II, a small number of teams will implement elements of a new architecture, building on the insights gained in FIND Phase I. The plan is for this phase to start in FY09. Phase III will render select architectures into practice by fleshing out the details of protocols and producing code that could run on research infrastructures. The plan is for the third phase to start in FY12. Initial experiments, in Phase II and Phase III are likely to lead to iterations in design, so there will be continuing opportunities to propose and integrate new concepts into the most promising future architectures. New program announcements will articulate the kinds of research that will be included in Phase II and Phase III of FIND.

FIND Principal Investigators (PIs) are required to attend FIND PI meetings three times each year. The purpose of these meetings is to discuss and identify a set of requirements that a future Internet should be designed to meet, discover and learn about research that is relevant to FIND objectives, and build awareness and understanding toward the eventual emergence of integrated architectural proposals.

Scientific Foundations for Internet's Next Generation (SING)

The SING Program invites proposals in computing, communications, signal processing, and network science that provide new foundations for a clean-slate approach to network design.

SING projects are expected to challenge well-established theories and break barriers between, for example, theoretical and experimental research communities, networking and computing researchers, and researchers focused on the physical and network layers. It also provides an opportunity for foundations researchers to validate the assumptions made in network models.

SING is currently supporting research across a broad range of topics including, but not limited to:

• Address scalability, complexity, and interactivity challenges;
• Develop signal processing techniques in support of content analysis and manipulating massive data sets;
• Understand tradeoffs among communication, computation and storage;
• Create new models for mobility-enhanced information dissemination; and
• Design signal processing, computing, and communications techniques for pervasive computing and communication environments.

Next-Generation Networked Information (NGNI)

The NGNI Program is encouraging research that can inform the development of both information systems and the networking technologies that underlie them.

NGNI seeks innovative research on information systems that anticipate future distributed networking environments. NGNI is supporting research across a broad range of topics including, but not limited to:

• Design new information systems for networks of extreme heterogeneity and complexity;
• Develop new metaphors and models that give coherence to the highly diverse ways users might access information in the future;
• Develop new applications that provide information based on both content and context; and
• Create new approaches for providing timely and coherent access to information when the magnitude and speed of the information flow dwarfs our ability to transport, process, or comprehend it directly.

NGNI projects should articulate the anticipated future networking environment, as well as the means for evaluating results. Projects that involve partnerships between researchers in Information and Intelligent Systems and Networking and Distributed Systems are especially encouraged.

Bridges to the Future

CISE is encouraging members of all network science and engineering research and education communities to collaborate in projects that will seed and sustain the transformation of networking research over the next decade. It is expected that advances will be made when PIs doing networking research, for example, learn more about the new applications that other computer scientists are imagining and would like to develop – if only the network were designed differently. Similarly, we expect that collaborations between theory and networking researchers will result in increased relevance and validity of foundational theories and models as well as new knowledge about how our current networks work. CISE is asking all members of its research and education communities to look beyond their own sub-disciplines and enter into new collaborations that will advance networking over the next decade.

During FY08, CISE will host informational meetings and workshops to bring the PIs of the FIND, SING, and NGNI programs together with other interested PIs and their Program Directors to consider how best to encourage broad, interdisciplinary research collaborations that promise to transform network research over the next decade.

CISE is also encouraging the community to experiment and explore radical network designs and policies in existing testbeds, such as PlanetLab, Emulab, and other experimental network infrastructures. The expectation is that today's investments will produce ambitious and innovative projects, which will bridge the gap between what is currently available for network experimentation and what is possible in the future if we were to have large-scale (i.e., multi-national or global), heterogeneous (i.e., with many underlying technologies), federated (i.e., connecting many complementary, but independent facilities), programmable network infrastructures with real users.