We propose a new internetworking architecture that represents a departure from current philosophy and practice, as a contribution to the ongoing debate regarding the future Internet. Building upon our experience with the design and prototyping of the just-in-time protocol suite, we outline a framework consisting of (1) building blocks of fine-grain functionality, (2) explicit support for combining elemental blocks to accomplish highly configurable complex communication tasks, and (3) control elements to facilitate (what is currently referred to as) cross-layer interactions. In this position paper, we take a holistic view of network design, allowing applications to work synergistically with the network architecture and physical layers to select the most appropriate functional blocks and tune their behavior so as to meet the application's needs within resource availability constraints. The proposed architecture is flexible and extensible so as to foster innovation and accommodate change, it supports a unified Internet, it allows for the integration of security and management features at any point in (what is now referred to as) the networking stack, and it is positioned to take advantage of hardware-based performance-enhancing techniques.

While research on cross-layer network optimization has been progressing, useful implementations have been lagging because the current Internet architecture does not accommodate cross-layering gracefully. As part of our FIND project, we propose a software architecture for the future Internet that is designed to accommodate such interactions. We present a conceptual overview as well as high level software design and an early prototype implementation, and point out the strengths of our architecture.

Optical burst switching (OBS) is emerging as an optical networking technology combining best of the features of optical circuit-switching technology and optical packet-switching technology. This paper discusses architecture and hardware implementation of a network interface card (NIC) for an OBS network. The NIC implements a specific low-latency signaling protocol just-in-time (JlT) for OBS networks. This NIC can handle data streams up to 1 Gbps. Average OBS JIT signaling message processing time with this NIC is on the order of 1 microsecond. This is the first JIT OBS NIC implementation known to the authors. Complementary efforts are on-going to develop other elements for the OBS network.

This paper studies the important fault management issue with focus on the fast restoration mechanisms for optical burst switched (OBS) networks. In order to reduce the burst losses during the restoration process, effective fast restoration schemes are necessary. This is illustrated via two basic fast restoration schemes, the distributed deflection scheme and the local deflection scheme, compared with the slow global routing update mechanism. A novel priority-based QoS restoration scheme is also proposed to provide differentiated restoration services. Through detailed descriptive analysis and a comprehensive simulation study, these fast restoration schemes demonstrate fast restoration process, low fault management overheads, and excellent burst loss performance. As far as we know, this is the first comprehensive study on the restoration mechanisms for OBS networks.

We describe: (i) the architecture of an optical burst switched (OBS) demonstration network overlaying the ATDnet transparent all-optical testbed, and (ii) experiments underway in the testbed. The OBS overlay uses a simple hardware-based protocol embedded in OBS network controllers to manage commercial off-the-shelf DWDM switches. Data paths are all-optical and completely transparent, and can carry analog or digital traffic in any format, data rate, and modulation scheme. Experiments with latency- and jitter-sensitive HDTV transmission, petabyte file transfers, and immersive real time visualization of satellite imagery over the OBS network are ongoing. Parallel research on transport protocols, QoS-aware routing protocols, adaptors for an OBS LAN, and network management architecture are applied as completed. This is the first just-in-time (JIT) OBS field trial known to the authors.

The paper describes an experimental key agile cryptographic system under design at MCNC. The system is compatible with ATM local- and wide-area networks. The system establishes and manages secure connections between hosts in a manner which is transparent to the end users and compatible with existing public network standards. A Cryptographic Unit supports hardware encryption and decryption at the ATM protocol layer. The system is SONET compatible and operates full duplex at the OC-12c rate (622 Mbps). Separate encryption keys are negotiated for each secure connection. Each Cryptographic Unit can manage more than 65,000 active secure connections. The Cryptographic Unit can be connected either in a security gateway mode referred to as a 'bump-in-the-fiber' or as a direct ATM host interface. Authentication and access control are implemented through a certificate-based system. The current status of the system is that hardware and software detail designs have been completed. An early version of the key management software has been completed and demonstrated. Hardware fabrication and systems integration are expected to take place over the next several months. Once completed the proof-of concept system will be used to explore issues of privacy, access control and authentication in relation to communications over emerging public networks.>

The VISTAnet gigabit testbed project features the deployment of a network connecting distant supercomputers and research programs in communications, advanced graphics techniques, and a medical application. It has been established to experiment with and evaluate techniques associated with gigabit networking. The testbed goals, infrastructure, and driving applications are described. Early insights from the testbed activities are presented to identify important problems associated with networked supercomputers and their impact on public broadband integrated services digital networks. The conflict between the performance needs of supercomputer communications and the constraints of cell-oriented communications are identified and discussed. The concepts of metacomputing are introduced along with their network implications. The metacomputing model is applied to the VISTAnet application to illustrate these points.>