Ken Carlberg

Building shared trees using a one-to-many joining mechanism

This paper presents a new approach for building shared trees which have the capability of providing multiple routes from the joining node onto an existing tree. The approach follows a design parameter of CBT and PIM in that it operates independently of any unicast routing protocol. However, a paradigm shift is introduced such that trees are built in an on-demand basis through the use of a one-to-many joining mechanism. In addition, the paper presents optimisations of the new mechanism to help constrain its impact in the case where many receivers exist for a given multicast group.

Lower than best effort: a design and implementation

In recent years, the Internet architecture has been augmented so that Better-than-Best-Effort (BBF) services, in the form of reserved resources for specific flows, can be provided by the network. To date, this has been realized through two different and sequentially developed efforts. The first is known as Integrated Services and focuses on specific bounds on bandwidth and/or delay for specific flows. The Differential Service model was later introduced, which presented a more aggregated and local perspective regarding the forwarding of traffic. A direction that is missing in todays work on service models is a defined schema used to purposely degrade certain traffic to various levels below that of Best Effort. In a sense, a new direction that provides a balancing effect in the deployment of BBE service. This is particularly evident with continual and parallel short transaction flows (like that used for web applications) over low bandwidth links that are not subject to any backoff penalty incurred by congestion because state does not persist. In a more indirect perspective, our model correlates degraded service with the application of usage and security policies — administrative decisions that can operate in tandem or disjointly from conditions of the network. This paper attempts to address these and other issues and presents the design and implementation of such a new degraded service model and queuing mechanism used to support it.

DFLOW: Low latency congestion control

This paper provides an overview of the DFlow congestion control algorithm, which aims to provide for lower delay and lower loss media transport. We provide an evaluation of the algorithm in a simulator which shows that it can provide for self-fairness and low delay operation. Furthermore we demonstrate that it can maintain reasonable throughput against TCP Vegas and LEDBAT.

The RadioActive networking architecture

This paper describes the activities in application level active networks (ALAN) under the DARPA-funded RADIOACTIVE Project; this is closely related to work carried out under a European Commission project ANDROID. The ALAN infrastructure was developed mainly under other projects; it is summarized mainly for background. The version used relies on separate active applications driven by policies, with the policies expressed in XML. There are two principal applications carried through in this project: adaptation of multicast, multimedia conferencing tools (M-Bone) and Virtual Private Networks (VPNs). The former were developed in other projects; the latter derives from the X-Bone overlay networks of ISI. It is an important aspect of the project that the final activity all works in the context of IPv6. The paper describes the measures that were required to make the applications into ALAN ones, and the problems encountered in moving all the components to work in the IPv6 environment.

Media Transmission over Coupled Wired/Wireless Networks using Application Level Active IPv6 Networks

A general architecture for Edge services in the IPv6 environment has been developed in several active networks projects. In a series of multimedia, multicast, conferencing projects, it became clear that an edge media adaptation service is needed if the participating networks are heterogeneous. This need became more acute with wireless access to the Internet. This paper describes both the general architecture, and the functionality of the components, to achieve active edge services. It considers also how the IPv6 functionality assists in setting up a generic service, which may be of interest to wireless operators.