We describe the design and performance of storage area network multimedia server (SAM), which manages a set of parallel storage devices connected by a storage area network. SAM is geared towards providing multiple concurrent services such as interactive video-on-demand and video games. The challenge, therefore, is twofold: provide a high throughput in order to service the maximum number of users possible, with a small response time that is necessary for interactive applications. The primary focus of this paper is on the data layout scheme for SAM. The problem of out-of-order delivery of video frames is described, and a scheme to transmit the frames in order is presented. A data layout scheme called constrained random striping (CRS) is developed for the logical loop model. Using a simulation model, the performance of this scheme is compared with several other data layout strategies. It is shown that CRS reduces the start-up latency of clients. The performance gains and tradeoffs of the different strategies are discussed in detail.

Social networks have first been studied by social scientists to understand the patterns of relationships between people. More recently, social networking on the Internet capitalizes on the Web’s latent structure as a meta-network of social connections to boost computer-supported collaboration in conjunction with the use of Semantic Web metadata. Such metadata are generally machine-created and per se lack certification or trust grading. This paper examines how social network information can be used to accelerate reliable metadata creation. In particular, we focus on the application of Web annotations to leverage asynchronous collaboration through the Semantic Web by introducing the notion of a user’s Web personality to model the various trust levels in collaboration.

Increasing sophistication of Internet services requires more refined methods of service differentiation in the delivery of data and media flows. Earlier work on quality-of-service (QoS) routing has focused on the network mechanics to find the best path for admitted connections under specific QoS constraints, and to maximize global resource utilization. We look at the related notion of qualification-based routing, where routers check on the fulfillment of one or more constraints before handling a particular packet flow. Packets are demarcated and evaluated by qualifiers in their headers, which may indicate for example an authorization level in secure transmissions, or a payment receipt for a media-on-demand application. In particular, this paper explores the coordinative processes for tree formation and repair in qualification-based multicast routing, with the goal to leverage group-centric communication for future mass media Internet applications. We introduce the QuORRUM protocol to implement qualifier-driven sparse-mode tree building and repair and show that it performs better than other protocols in terms of bandwidth usage and repair times

The amount of information requested over the World Wide Web has increased enormously during the past decade. Web caching helps to reduce service times, balance the load to origin servers, and brings content closer to the user. Since duplicating and distributing files amongst proxy caches has proved to be insufficient, cooperative caching aims to ameliorate the shortcomings of basic replication caching through inter-cache communication. This paper categorizes and evaluates cooperative caching schemes with regard to user access patterns from established traffic archives. We use cache statistics at the proxy level and discuss possible causes for cache misses with the aim to bring greater understanding to the interplay of access and caching for web documents.

We discuss the concept of adaptive middleware for augmented living spaces, with the aim to provide a framework to systematically outfit homes with the flexible intelligence to cooperatively optimize their shared resources. To exemplify this concept, we discuss the EMISS testbed and platform for rapid design, simulation and deployment of wireless sensor networks for the "adaptive home" beyond off-the-shelf control mechanisms for networked home appliances. Based on a combination of web and object technology, the system offers location-independent, user-friendly access to customized energy-aware home control, as well as the basic ambient intelligence to make a home adapt to user behavior and environmental constraints.

Collaboration infrastructures are gradually coming of age, supporting real-time audio and video with a mixture of network core and edge based techniques to enable large-scale, multi-party communication. User participation in collaborative activities requires mechanisms to support dynamic group membership and maintenance of perpetually changing delivery topologies. In the past, multicast support has been studied within testbeds such as the MBone, and across overlay architectures. In this paper we tackle the problem of how network-native multicast can help to orchestrate the formation of virtual teams in group collaboration. We argue that user interaction can be understood as a series of constraint-driven information exchanges, for example by sharing objects based on access rights. A novel multicast protocol, Limited Core Fix (LCF), is introduced to efficiently maintain multicast delivery trees among distributed work teams. LCF is a receiver-centric mechanism to support qualification-based multicast routing and performs more efficiently than comparable existing protocols.

Distributed media synchronization is of ever increasing importance in the wake of an exponentially growing number of networked devices, media on demand, and user interactivity. Many protocols have been proposed to synchronize delivery of multimedia data among user groups spread over a packet network. These protocols typically focus on multimedia synchronization in small groups, may not support multimedia sessions with multiparty turn-taking patterns, are often application-specific, and do not perform well with varying network topologies. We describe the multipoint synchronization protocol (MSP) as a time coordinating mechanism that works well in a variety of group settings and for a variety of applications and network conditions, using multicast support if available. MSP is fully distributed and designed to scale to large user groups supporting turn-taking or concurrent sending and receiving. It has no central point of failure and can synchronize concurrent media streams over best-effort packet delivery networks in the absence of a global physical clock.

The majority of today's Internet applications relies on point-to-point communication. In recent years, however, multipoint communication support has become the foundation for such applications as multiparty video conferencing, distributed interactive simulations, and collaborative systems. We describe a novel protocol to coordinate multipoint groupwork within the IP-multicast framework. The protocol supports Internet-wide coordination for large and highly-interactive groupwork, relying on the dissemination of coordination directives among group members across a shared end-to-end multicast tree. We also describe how addressing extensions to IP multicast can be used for our multisite coordination mechanism.

This paper discusses a group coordination architecture to support Internet-wide distributed collaboration in the context of legacy Internet protocols. Group coordination in distributed systems and multimedia systems has many faces manifested in a variety of user interfaces and network protocols. To date, no standardized methodology for engineering group coordination protocols exists. We perceive coordination as the third complementary component in the trinity of group-communication services, next to membership and dissemination. With the current surge in e-commerce and Web-leveraged information exchange among users, the need for systems offering better telepresence and interaction capabilities becomes tangible. Services to support distributed group interaction at (near) real-time, with user-specified quality-of-service, and at Internet scope are of particular interest in this mosaic of telepresence and remote collaboration. We propose a general group coordination architecture for heterogeneous networks, as a framework to leverage the rapid development of group-oriented distributed collaborative applications in the Internet, for example for distance education, distributed scientific simulation or visualization, and similar applications.

: Recent advances in computer hardware and networking technology have incited the deployment of wide-area streaming media services in the Internet. While such efforts as video-on-demand are largely limited to unidirectional delivery of content to the desktop, synchronously interactive group-oriented application services are foreseeable. In such applications, users collaborate on a shared workspace and freely exchange information in real-time under the premise of coordination and conflict freedom. Telecollaborative applications such as telemedicine or distance learning may profit from such coordination services. Ultimately, group coordination allows for groupware-style computing at Internet scope. The current IP-multicast framework contains provisions for group membership control and reliable dissemination services, however, it lacks support for group coordination. In this paper, we present a framework on network control and coordination functions to orchestrate synchronous multimedia groupwork. Our goal is to achieve a better understanding of the group coordination problem as an important component of future Internet multimedia collaboration tools.

The majority of today's Internet applications relies on point-to-point transmission. In recent years, however, multicast transmission has become the foundation for such applications as multiparty video conferencing, distributed interactive simulations, and collaborative systems. We describe a novel protocol to coordinate multipoint groupwork in the IP-multicast framework. The protocol supports Internet-wide coordination for large and highly-interactive groupwork, relying on transmission of coordination directives between group members across a shared end-to-end multicast tree. We also describe how addressing extensions to IP multicast can be put to use for our multisite coor-

Current research on networked multimedia applications such as distributed interactive simulations concentrates on transport issues like multicast routing and presentation and session management, including session orchestration and quality-of-service support in media delivery. Session orchestration includes mechanisms to coordinate fair and exclusive access to shared resources whose semantics do not allow for concurrent usage, in order to prevent conflicts and inconsistencies in the shared workspace. The authors discuss the current state of research with regard to such services, which they refer to as group coordination, and present a new approach that integrates group coordination with extended multicast services

Improvements in networking allow for increasingly complex collaboration environments with regard to sessions scale, range of shared tasks, and distance between remote parties. Floor control protocols add an access discipline to such environments that allows to mitigate race conditions on shared resources and throttle media transmission. Primary causes for resource competition among users may be the lack of mutual awareness sand formal session orchestration, or network and shot limitations. Various, often proprietary and unscalable solutions for floor control have been implemented for telemedicine, video conferencing, or distributed interactive simulation. To this date, an analytic comparison of the efficacy of these solutions is lacking. With efficacy, we mean the proportion of time that a protocol takes to allocate a resource, accounting for social and technical overhead for muser behavior, protocol cost, and network conditions. We present a novel taxonomy an comparative performance analysis of known classes of floor control protocols, including socially driven protocols, collision sensing on shared resource, floor taken passing in fully-connected and ring topologies, and, innovatively, across shared control trees. Accordingly, aggregated and selective transmission of control information over a multicast control tree offers the best scalability and efficacy. A novel hierarchical floor control protocol correlating in its operation with tree-based reliable multicast is outlined.

Group collaboration in distributed multimedia environments extends gradually to larger groups and wide area networks. While reliable multicasting has made significant advancements in recent years, effective mechanisms to synchronize and coordinate work within large multicast groups and across long distances are still lacking. Group coordination is here understood as the mediated access to shared remote resources in synchronous groupwork, as for example in telecollaboration and distributed simulation environments, complementing protocols for group membership, media synchronization and reliable ordered multicast. A comparative analytic model for known classes of group coordination mechanisms, ranging from socially mediated control to floor control in ring and tree topologies, is presented. It is shown that hierarchical group coordination is the most efficient and scalable approach to date. Based on these findings, a novel protocol is described, which dynamically organizes participants in a multilevel control tree and aggregates resource sharing directives on the paths between interacting stations.

The effectiveness of collaborative multimedia systems depends on the regulation of access to their shared resources, such as continuous media or instruments used concurrently by multiple parties. Existing applications use only simple protocols to mediate such resource contention. Their cooperative rules follow a strict agenda and are largely application-specific. The inherent problem of floor control lacks a systematic methodology. This paper presents a general model on floor control for correct, scalable, fine-grained and fair resource sharing that integrates user interaction with network conditions, and adaptation to various media types. The motion of turn-taking known from psycholinguistics in studies on discourse structure is adapted for this framework. Viewed as a computational analogy to speech communication, online collaboration revolves around dynamically allocated access permissions called floors. The control semantics of floors derives from concurrently control methodology. An explicit specification and verification of a novel distributed Floor Control Protocol are presented. Hosts assume sharing roles that allow for efficient dissemination of control information, agreeing on a floor holder which is granted mutually exclusive access to a resource. Performance analytic aspects of floor control protocols are also briefly discussed.

Floor control allows users of networked multimedia applications to utilize and share resources such as remote devices, distributed data sets, telepointers, or continuous media such as video and audio without access conflicts. Floors are temporary permissions granted dynamically to collaborating users in order to mitigate race conditions and guarantee mutually exclusive resource usage.A general framework for floor control is presented. Collaborative environments are characterized and the requirements for realization of floor control will be identified. The differences to session control, as well as concurrency control and access control are elicited. Based upon a brief taxonomy of collaboration-relevant parameters, system design issues for floor control are discussed. Floor control mechanisms are discerned from service policies and principal architectures of collaborative systems are compared. The structure of control packets and an application programmer's interface are proposed and further implementation aspects are elaborated. User-related aspects such as floor presentation, assignment, and the timely stages of floor-controlled interaction in relation to user-interface design are also presented.