We propose a hybrid approach for conference applications for Named-Date Networks (NDN). This hybrid approach combines a participant-driven, distributed server-based approach for conference control and management with a server-less approach for media forwarding. The participant-driven, distributed server-based approach provides robust and efficient conference control and management that is resilient to single point of failure and can scale dynamically with any conference size. The server-less media forwarding eliminates traffic concentration and congestion in the network. We have carried out experiments using XMPP-based conference tools over CCNx and have successfully built the media forwarding for two conference tools (audio and whiteboard) on Android platform. This hybrid approach leverages NDN to support robust and efficient conference applications.

Architectures for Internet quality of service (QoS) have been under discussion for over a decade and, with the commercialization of the Internet, the topic has become increasingly important. This paper gives a background and history of QoS for the Internet, then introduces and motivates the differentiated services (Diffserv) approach. The major advantages of the Diffserv approach are that it is a good match to the Internet architecture and that it can be initially deployed with a minimalist approach, adding complexity as needed. Despite the long history of discussion, the phrase "quality of service" does not have a universally accepted meaning. In this paper QoS is used to describe a set of measurable parameters, such as delay, throughput, and loss rate, that can be attached to some identifiable subset of the traffic of IP packets through a given network domain. The identifiable subset of traffic belongs to a "user" of IP QoS where "user" spans a range of granularities, from a single application program to an entire company. Providing guarantees about the values of network parameters requires the implementation and deployment of physical mechanisms throughout the network and then configuring these mechanisms in such a way that their effect, when viewed from the edges of the network, composes into the desired QoS. Diffserv uses simple mechanisms in a more complex composition, allowing the details of the composition to evolve while the mechanisms, part of the network infrastructure, can remain the same. The paper discusses the specifics of this approach and why it is well-matched to the Internet. Some practical issues for deployment are addressed. Further we address resource allocation and configuration questions, including simple possibilities for early deployment and some of the open questions for a more complex future deployment. This paper takes the position that it is possible to maintain reasonable QoS levels without recourse to any of the class of constrained routing approaches (including MPLS), though Diffserv can be used with these approaches if desired.

Since the publication of the article by Leland, Taqqu, Willinger and Wilson (1993) it has become well known that real network traffic does not follow a Poisson process and that feedback mechanisms in a network affect the traffic patterns. Yet much modeling and simulation work employs exponential interarrivals in open-loop traffic sources. Historically these sources were chosen because they are amenable to analysis and easy to simulate, but they have serious shortcomings. Even when the source distribution is changed to reflect the burstiness of real traffic, we found that open-loop models fail to reproduce the detailed time structure and the crucial effects of feedback on traffic sources. An evaluation of a hybrid fiber coax (HFC) architecture is used to illustrate the inadequacy of traditional open-loop modeling. In addition, a promising intermediate technique of using "local feedback" is shown. Finally using feedback models, the interaction of the HFC architecture with TCP dynamics was studied using Web-browsing and file transfer applications. These studies led to a novel architectural improvement that would have been missed in an open-loop model.

The authors describe a performance study of a trial switched multimegabit data service (SMDS) link (intended for inter-LAN connection) from the perspective of customers evaluating the feasibility of the link for some target applications. The goals were to take all measurements on the customer premises and to develop a methodology general enough to be used by customers to evaluate the link. The authors measured a lightly loaded system and developed a model of the SMDS connection suitable for evaluating applications via analysis or simulation. They document their methodology and present the SMDS connection delay values as well as a likely breakdown of the constituents of that delay. They used these data to create a simulation model and to simulate a simple application. In the trial configuration, where geographical distances were small, SMDS network delay was one of the notable components of end-to-end delay in the SMDS connection. However, for most packets, throughput is limited by the T1 capacity for transmitting SMDS cells, not by the SMDS network capacity.>

PARET (Parallel Architecture Research and Evaluation Tool) is an object-oriented graphical environment for the study of parallel or multiprocessor systems that provides a laboratory for investigation of alternative architectural decisions and experiments with such issues as scheduling, mapping, and routing policies. PARET is unique in providing an animated interactive visual tool for parallel systems with a modeling paradigm that is closely linked to its graphical representation. Models have the same morphology regardless of the subsystem represented or the level of detail in the model. Presently, PARET runs on SUN workstations, using color graphics, mouse input, and pop-up menus to control and observe a simulation engine. Ongoing projects using PARET include evaluation of a special-purpose multiprocessor system and use of PARET in a system to create application-specific ICs for signal processing.>

Special-purpose networks of processors and intelligent switching devices can be programmed to solve problems with inherent parallelism. The target algorithm is coded into a number of asynchronously executing parallel tasks, and assigned to specific processors. For well-known algorithms, information about the data communicated between these tasks (traffic) can be characterized and used by a heuristic algorithm to produce a specialized interconnection network of switching devices at a lower cost and increased performance over the use of a general purpose structure. A program was developed which uses intertask communications to automatically generate interconnection structures of switch chips (SC's). Simulation results were used to compare cost/performance measures for the program-generated networks with some well-known structures, using both statistically generated traffic patterns and the simulation of real applications.