Ip Multicast

Already a popular application in the Internet, IPTV is becoming, among the service providers, a preferred alternative to conventional broadcasting technologies. Since many of the existing deployments have been done within the safe harbor of telco-owned networks, IP multicast has been the desired streaming solution. However, previous studies showed that the popularity of the TV channels follows the Pareto principle, with the bulk of TV channels being watched only by a small fraction of viewers. Recognizing the potential scalability issues, we believe that multicast streaming approach may not be desirable for unpopular TV channels, especially when there are many such channels in the provider's service package. For this reason, the peer-to-peer content distribution paradigm is seen as an alternative, in particular for non-popular content. In order to analyze its viability, in this paper we perform a comparative analysis between IP multicast and a peer-to-peer overlay using unicast connections as streaming approaches, in the context of channels with different degrees of popularity. The analysis targets the bandwidth utilization, video quality and scalability issues, and our findings show that while multicast is always more efficient, peer-to-peer has a comparable performance for unpopular channels with a low number of viewers.

IP multicast is nowadays a common network technique for communication and data delivery between senders and a large set of receivers. It could be argued that multicast is not generally supported by ISPs yet it became a standard for media streaming, distributed computing, financial applications and many others. There is a need for suitable modeling and simulation tools supporting multicast routing, distribution trees and visualization of multicast data flows. In this paper we report on our so called ANSA extension which expands INET framework in OMNeT++ environment.

The bulk power system faces many challenges as load growth continues to far outstrip new transmission capacity, utility operators are retiring in large numbers, and renewable sources of energy and microgrids with very different power characteristics must be integrated into electricity grids and be actively managed. NASPInet holds great promise to help improve the reliability, efficiency, and cyber-security of the bulk power system. In order to achieve this potential, NASPInet must support a wide range of guarantees for latency, rate, and availability. It also must support delivery with extremely stringent values of low latency and high availability, and do so across a wide area even in the face of potential IT failures and cyber-attacks. In this paper, we describe the baseline delivery requirements for performance and fault-tolerance that NASPInet must provide if it is to meet those goals. We then offer twenty implementation guidelines that we argue must be met if the delivery requirements are to be implementable and at reasonable cost. These guidelines are based on a number of sources, including our 11 years of designing and developing GridStat (an implementation of the NASPInet Data Bus), the experience of us and others on projects by DARPA and others involving widearea, real-time, fault-tolerant, and secure middleware for widearea networks, and the state of the art and practice in networking and distributed computing. Finally, we summarize the coverage of those delivery requirements and implementation guidelines that middleware, networking protocols IP Multicast and MPLS, and power protocols C37.118 and IEC 61850, provide.

In this paper we describe SelectCast, a self-repairing multicast overlay routing facility for supporting publish/subscribe applications. SelectCast is a peer-to-peer protocol, and leverages Astrolabe, a secure distributed information management system. SelectCast uses replication to recover quickly from transient failures, as well as Astrolabe's aggregation facilities to recover from long-term failures or adapt to changes in load or QoS requirements. In order to evaluate the scalability and performance of SelectCast, and compare these with other multicast facilities, we built a multicast testing facility on NetBed. This paper reports latency and load results for Se-lectCast, compared to both native IP multicast and Yoid.

This article presents the functioning of IP multicast working on MPLS for the traffic without quality of service requirements through the hop-by-hop method extended for multicast and for the traffic with quality of services (QoS) requirements through reserve of resources with traffic engineering protocols extended the signaling for multicast such as CR-LDP and RSVP-TE. Finally, the behavior of both methods is showed through simulation by analyzing the delay like parameter of QoS when a multicast of video stream or videoconference is carried out.

This paper proposes a novel approach centered on multimedia internetworking for the development of distributed virtual instruments. Multimedia internetworking refers to network infrastructures, protocols, models, applications and techniques being currently deployed over the Internet to support multimedia applications, e.g., videoconferencing, video-on-demand, shared workspaces. It is applied to broaden the concept of virtual instrument and enable new measurement patterns leveraging efficiency and interactivity. A Distributed Virtual Instrument (DVI) is a virtual instrument split into possibly multiple and independent parts, sender and receiver, which are linked by real-time continuous media and control streams. Senders and receivers are built by using open, composable and modular components based on a time sensitive actor framework and glued by multimedia middleware. A prototype is described to demonstrate the potential and the benefits of the proposed approach.

Giancarlo Fortino, Angelo Furfaro, Wilma Russo DEIS, Università della Calabria Via P. Bucci s/n, 87036 Rende (CS), Italy E-mail: {g.fortino, w.russo, a.furfaro}@unical.it ... Juan C. Guerri, Ana Pajares, Carlos E. Palau Universidad Politecnica de Valencia Camino de ...

In recent years, overlay networks have become an effective alternative to IP multicast for efficient point to multipoint communication across the Internet. Typically, nodes self-organize with the goal of forming an efficient overlay tree, one that meets performance targets without placing undue burden on the underlying network. In this paper, we target high-bandwidth data distribution from a single source to a large number of receivers. Applications include large-file transfers and real-time multimedia streaming. For these applications, we argue that an overlay mesh, rather than a tree, can deliver fundamentally higher bandwidth and reliability relative to typical tree structures. This paper presents Bullet, a scalable and distributed algorithm that enables nodes spread across the Internet to self-organize into a high bandwidth overlay mesh. We construct Bullet around the insight that data should be distributed in a disjoint manner to strategic points in the network. Individual Bullet receivers are then responsible for locating and retrieving the data from multiple points in parallel. Key contributions of this work include: i) an algorithm that sends data to different points in the overlay such that any data object is equally likely to appear at any node, ii) a scalable and decentralized algorithm that allows nodes to locate and recover missing data items, and iii) a complete implementation and evaluation of Bullet running across the Internet and in a large-scale emulation environment reveals up to a factor two bandwidth improvements under a variety of circumstances. In addition, we find that, relative to tree-based solutions, Bullet reduces the need to perform expensive bandwidth probing. In a tree, it is critical that a node's parent delivers a high rate of application data to each child. In Bullet however, nodes simultaneously receive data from multiple sources in parallel, making it less important to locate any single source capable of sustaining a high transmission rate.

This paper presents a scalable archite ctur efor supporting large-scale inter active Internet games. In order to support a large number of particip antsand to divide the workload, the virtual world is divided into partitions. Each partition is then assigned to a server. A client (i.e., a player or a particip ant)will join a server ac c ording to the position of the avatar it controls. Compared t o a c entr alized a r chite ctur e, this distribute d client-server archite ctur e i s m o r e s c alable. In addition, compared to a fully distribute d,peer-to-p eerarchite cture, it also provides a means for dete cting cheating in distribute dgames. Sinc einteractions and accounting information must be forwarded directly to one of the servers for quali cation and veri c ation,cheating amongst distribute dplayers of the game will be minimized. T o support secur ed c ommunication for interactions and accounting information as well as to speedup periodic update messages (e.g., position updates), a hybrid communication scheme using both TCP and IP multicast is used betwe enclients and the asso ciate d server. The communication among servers is enabled by the R un-TimeInfr astructure (R TI)services. The High Level Archite ctur e (HLA) Data Distribution Management (DDM) is employed to limit the amount of communication betwe enthe servers. In addition, the Ownership Management (OM) is also employed to implement the ne edfor transferring the avatars betwe en servers. In this p ap er, the design detail of the ar chite cture

This article outlines the challenges of IP multicast over Mobile IP. Then it provides a comprehensive overview of existing multicast solutions to handle mobile sources and receivers in both the Mobile IPv4 and the Mobile IPv6 environments. The solutions are classified into different classes in light of how the network infrastructure is used and the nature of the multicast members (receivers or sources). For each solution, we present a brief overview; we describe the architecture and the proposed protocol; we discuss the advantages and the limitations; and we compare qualitatively all the solutions of each class based on common criteria such as optimal routing, join latency, handover transparency, etc.

The conventional wisdom has been that IP is the natural protocol layer for implementing multicast related functionality. However, more than a decade after its initial proposal, IP Multicast is still plagued with concerns pertaining to scalability, network management, deployment and support for higher layer functionality such as error, flow and congestion control. In this paper, we explore an alternative architecture that we term End System Multicast, where end systems implement all multicast related functionality including membership management and packet replication. This shifting of multicast support from routers to end systems has the potential to address most problems associated with IP Multicast. However, the key concern is the performance penalty associated with such a model. In particular, End System Multicast introduces duplicate packets on physical links and incurs larger end-to-end delays than IP Multicast.

Real-time network applications need a special attention. Mechanisms allowing them to face delay variation on the flow delivery must be available as much as resource reservation and admission control for QOS guarantee of all the flows being accepted within the network. This paper presents an integrated stack of protocols enabling real-time network programming. The programming issues held by IP multicast, RTP, RTCP and RSVP protocols are discussed in details and tackled in both Sun Solaris and MS Windows NT environments.

Providing authentication mechanisms for IP-Multicast streams is paramount for the development of large scale commercial multicast content delivery applications. This need is particularly strong for the delivery of real time content, such as live video/audio news events or financial stock quote distribution. However, this turns out to be a quite challenging problem for many reasons. First, the authentication of the multicast data must be verifiable by a potentially very large number of untrusted recipients. Second, since multicast communication protocols are almost always best effort, the authentication mechanisms needs to authenticate received content despite the potential loss of some packets. Finally, the authentication mechanism needs to be efficient enough to cope with real time data and should have a small communication overhead. We propose a new multicast authentication scheme designed to authenticate real time multicast packet streams with a potentially unlimited number of re...

This paper proposes the modeling and the performance analysis of a high-level control protocol -COCOP, which enables cooperative groups of clients to control a shared server delivering time-dependant data services. Several synchronous multimedia systems such as media on-demand, web casting, and networks of real/virtual sensors, can beneficially exploit COCOP to furnish cooperative control sessions. The protocol can be mapped onto a multicast transport support based either on IP-multicast or on an application level multicast infrastructure. An event-driven simulation framework is purposely customized and exploited to analyze the protocol performance and thus the dynamics of a cooperative control session over simple, yet representative topologies.

The Internet research community has proposed many routing protocols to support efficient multicast application. However, these protocols were developed for multicast parties whose members are topologically stationary, and they thus do not support both mobile receivers and mobile sources. Unfortunately, the mobility of the multicast source is a challenging problem to solve. When a multicast source moves from one IP network to another, it obtains a new IP address. However, since IP multicast is receiver-initiated, and the multicast delivery tree is constructed based on the multicast source’s address, the multicast traffic cannot be delivered to multicast receiver unless a new delivery tree is built. As a consequence, the multicast routing entities require new multicast routing states and multicast receivers need to be notified whenever the handover occurs. This induces a non-transparent mobility, and leads to session disruption. To overcome these drawbacks, this paper presents a new solution to make multicast source mobility transparent to both multicast receivers and on-tree multicast routers. This solution, called Mobility-aware Rendezvous Point, extends the PIM-SM Rendezvous Point with mobility awareness, and introduces new techniques for the source registration with the Rendezvous Point. Simulation results have proved that the solution avoids the reconstruction of a multicast tree after the handover of the multicast sources.

In recent years, IPTV has received an increasing amount of interest from the industry, commercial providers and the research community, alike. In this context, standardization bodies, such as ETSI and ITU-T, are specifying the architecture of IPTV systems based on IP multicast. An interesting alternative to support the IPTV service delivery relies on the peer-to-peer (P2P) paradigm to distribute and push the streaming effort towards the network edge. However, while P2P IPTV was studied in fixed access technologies, there has been little attention paid to the implications arising in mobile environments. One of these involves the service handover when the user moves to a different network. By analyzing previous work from the perspective of an IPTV service, we concluded that a proactive approach is necessary for the handling of inter-network handovers. In this paper, we propose a new general handover mechanism for the IP Multimedia Subsystem (IMS), while studying its applicability to a P2P IPTV service. Our solution, called Proactive Context Transfer Service, incorporates the existing IEEE 802.21 technology in order to minimize the handover delay. The proposal is validated by comparing it against solutions derived from previous work.

The NTDR development team is currently implementing 1P Multicast capabilities to support activities in the Tactical Internet. Protocols and mechanisms designed for wire based networks do not always transfer efficiently to an RF media. This paper introduces the tradeoffs, and the solution arrived at, for distributing Multicast traffic efficiently in the NTDR network, It will also describe the protocol RGMP developed to support Multicast group memberships for the NTDR Intranet and for the external Internet inte~ace to IGMP.

While multicast services are becoming very attractive, their large deployment and commercial use is currently slowed down partly due to the lack of integrated management solutions for the components that participate to the operation of these services at various levels. Excellent standalone components exist today and are good candidates for the integration. Joined and interfaced with standard management platforms, they cover most of the functions retaled to multicast service monitoring. In this paper we present the lesulting architecture of one integration effort which combines two multicast management tools for topology monitoring, pre-event testing and in-situ monitoring. The proposed architecture is used for service level monitoring and data collection.

Multicast communication is a topic of intense study by the network research community. The IP Multicast service of the network layer doesn't provide the desired reliability to some multicast applications, and the interest towards approaches to reliable multicast communication has increased. In this work, we focus on the Server Replication method, wherein the data are replicated over a subset of the multicast-capable relaying hosts and retransmission requests from receivers are handled by the nearest Replicated Server. The problem of selecting the best subset of the multicast-capable relaying hosts to replicate the data is NP-Hard. We propose a hybrid metaheuristic to find near optimal solutions for this problem. This proposal is based on a hybrid version of the GRASP metaheuristic that incorporates data mining techniques. Experimental results show that our technique outperforms existing approaches.

This paper proposes a design for key management for secure multicast in hybrid satellite networks. Communication satellites offer an efficient way to extend IP multicast services for groups in wide-area networks. In order to be commercially viable, the multicast traffic should be accessible only to paying subscribers. Access control can be achieved by data encryption. This requires secure and efficient methods to generate, distribute and update the keys. Most current key management protocols do not scale well when applied to large dynamic groups in wide-area networks. This paper attempts to solve the above problem for groups in a hybrid network that is composed of terrestrial Ethernet LANs interconnected by ATM-based satellite channels. We investigate current group key management protocols, and design a framework for secure and scalable key management for the multicast routing architecture in the satellite network. The proposed framework is presented in detail, alongwith analysis and simulation results.

In this paper, we propose DINPeer middleware to overcome limitations in current peer-to-peer (P2P) overlay systems. DINPeer exploits a spiral-ring method to discover an inner ring with relative largest bandwidth to form a DINloop (Data-In-Network loop). DINPeer further integrates DINloop with P2P overlay network via node state and routing algorithm. The key features of DINPeer include using DINloop to replace a multicast rendezvous point and turning DINloop into a cache to achieve data persistency. Simulations show that DINPeer is able to optimize P2P communication in a number of ways, such as when the size of the DINloop is capped within a limit, it can achieve a better performance than native IP multicast and P2P overlay multicast systems.

In this paper we present imss: IP Multicast Shortcut Service for ATM networks. imss pursues the \shortcut" routing paradigm in order to exploit the underlying ATM QoS and native routing mechanisms in the optimal way. imss is built on top of congress (CONnection oriented Group-address RESolution Service). Congress is an e cient native ATM protocol for resolution and management of multicast group addresses in a large ATM cloud. Congress resolves multicast group addresses and maintains their membership for applications. It is not designed to handle the applications' data-exchange.

This paper discusses the importan t role of multicastin g in designing, implementing, and managing interactive multimedia distance learning systems. This is achieved in the context of IRI, an Interactive Remote Instruction system for distance learning built at Old Dominion University. IRI is an Internet-b ased system which integrates continuous multimedia, shared applicatio ns and a variety of multiuser collaborative utilities. In this paper, w e concentrate on the process architecture and dynamic multicast group handling as they pertain to managin g multimedia resources, and show how they support robustness and short response time to user actions. IRI uses raw IP multicastin g for audio and video streams and reliable multicasti ng for resource management and data sharing. The system is scalable (uses multicast for inter-process communication) and expandabl e (partition ed into a set of autonomo us but cooperating components).

This paper proposes an efficient protocol and associate algorithm for group key management in secure multicast. This protocol is based on a hierarchy approach in which the group is logically divided into subgroups. The group key is organized using member secrets assigned to each member and server secrets assigned to each subgroup, and the inverse value of the member secrets are also used to manage the group key when a member leaves. In this case, each member in a single subgroup needs to store the inverse values of the other members in that subgroup with the exception of its own. When a member joins the group, after updating the previous group key in the server, the new key is sent to all existing group members, and the inverse value of the new member is sent to subgroup members (where there is a join), by exploiting IP multicast. Most importantly, the server just sends the inverse value of the leaving member to the subgroups when a member leaves. Then, the group key is updated by each remaining member in the subgroups by using that inverse value. Consequently, the benefits are two-fold. First, only one key needs to be generated by the server at each event. Second, not only the computational overhead is reduced but also new key information can be multicast to all members simultaneously. This paper describes the details of our novel protocol and the related algorithm.

MITv is a proposition for interactive television to be deployed using an association of current distribution television systems (e.g., satellite DVB-S, cable DVB-T, TV over xDSL, etc.) and classical Internet infrastructure. MITv makes use of the natural association between broadcasted TV programs and associated interactive contents to manage replication of those contents over an ad-hoc infrastructure. The replication phase can be based on IP multicast over satellite service, providing a high performance and realistic way to convey the interactive content over the geographically distributed replication units. The proximity between replication units and end-users then allow a very high quality for interactive media presentation by way of an access network such as xDSL. The MITv system was entirely developed and tested in real conditions through a satellite/xDSL network emulator.

This article presents the functioning of IP multicast working on MPLS for the traffic without quality of service requirements through the hop-by-hop method extended for multicast and for the traffic with quality of services (QoS) requirements through reserve of resources with traffic engineering protocols extended the signaling for multicast such as CR-LDP and RSVP-TE. Finally, the behavior of both methods is showed through simulation by analyzing the delay like parameter of QoS when a multicast of video stream or videoconference is carried out.

This article presents the functioning of IP multicast working on MPLS for the traffic without quality of service requirements through the hop-by-hop method extended for multicast and for the traffic with quality of services (QoS) requirements through reserve of resources with traffic engineering protocols extended the signaling for multicast such as CR-LDP and RSVP-TE. Finally, the behavior of both methods is showed through simulation by analyzing the delay like parameter of QoS when a multicast of video stream or videoconference is carried out.

Providing authentication mechanisms for IP-Multicast streams is paramount for the development of large scale commercial multicast content delivery applications. This need is particularly strong for the delivery of real time content, such as live video/audio news events or financial stock quote distribution. However, this turns out to be a quite challenging problem for many reasons. First, the authentication of the multicast data must be verifiable by a potentially very large number of untrusted recipients. Second, since multicast communication protocols are almost always best effort, the authentication mechanisms needs to authenticate received content despite the potential loss of some packets. Finally, the authentication mechanism needs to be efficient enough to cope with real time data and should have a small communication overhead.

Network bottlenecks, firewalls, restrictions on IP Multicast availability and administrative policies have long prevented the use of multicast even where the fit seems obvious. Our belief is that this poses a dilemma for the cloud, because many cloud applications will require extensive replication or serve large sets of clients (hence need multicast), and yet the prevailing communication model is point-to-point. Moreover, it seems unlikely that any single solution can run in all settings and address all plausible application requirements. Here, we present Quilt, a system that automatically weaves a patchwork of multicast regions each running different protocols, creating an efficient and scalable wide-area overlay. By dynamically exploring the environment at and between end-hosts, Quilt clusters nodes into patches, selecting the best multicast protocol from a developer-provided set on a patch-by-patch basis and adapting as needed. Quilt orchestrates inter-patch forwarding to maximize reliability while minimizing duplication. This paper discusses and then evaluates the system.

Ricochet is a low-latency reliable multicast protocol designed for time-critical clustered applications. It uses IP Multicast to transmit data and recovers from packet loss in end-hosts using Lateral Error Correction (LEC), a novel repair mechanism in which XORs are exchanged between receivers and combined across overlapping groups. In datacenters and clusters, application needs frequently dictate large numbers of fine-grained overlapping multicast groups. Existing multicast reliability schemes scale poorly in such settings, providing latency of packet recovery that depends inversely on the data rate within a single group: the lower the data rate, the longer it takes to recover lost packets. LEC is insensitive to the rate of data in any one group and allows each node to split its bandwidth between hundreds to thousands of fine-grained multicast groups without sacrificing timely packet recovery. As a result, Ricochet provides developers with a scalable, reliable and fast multicast primitive to layer under high-level abstractions such as publish-subscribe, group communication and replicated service/object infrastructures. We evaluate Ricochet on a 64-node cluster with up to 1024 groups per node: under various loss rates, it recovers almost all packets using LEC in tens of milliseconds and the remainder with reactive traffic within 200 milliseconds.

Efficient multicast congestion control (MCC) is one of the critical components required to enable the IP multicast deployment over the Internet. Previously proposed MCC schemes can be categorized in two: single-rate or multi-rate. Single-rate schemes make all recipients get data at a common rate allowed by the slowest receiver, but are relatively simple. Multi-rate schemes allow of heterogeneous receive rates and thus provide better scalability, but rely heavily on frequent updates to group membership state in the routers. A recent work by Kwon and Byers, combined these two methods and provided a multi-rate scheme by means of single-rate schemes with relatively low complexity.

In the terrestrial digital TV systems broadcasting is employed, meaning that all TV programs are sent over all transmitters, also where there are no viewers. This is inefficient utilization of spectrum and transmitter equipment. Applying interactive multicasting over DV approach that would substantially reduce the spect to deliver a certain amount of TV programs be achieved by Dynamic single-frequency network formations, which can be implemented using PARPS (Packet and Resource Plan Scheduling). heterogeneous program selection model system of four coordinated transmitters, assumptions, IP multicasting over non-continuous transmission DSFN gives 1740% increase in multiuser system spectral efficiency (MSSE) in (users•bit/s)/Hz/site broadcasting over SFN.

We consider the problem of multirate network design with point-to-multipoint communications. We give a mathematical formulation for this problem. Using approximations, we show that traffic concentration on a small number of links significantly reduces the cost of the network. We then propose a heuristic based on the traffic concentration principle to solve the network design problem approximately. Because this heuristic no longer requires advanced knowledge of demand matrices, we explain how it can be used as the basis for real-time design procedures. By means of numerical results, we show that this heuristic yields nearly optimal solutions.

Large-scale applications are characterized by a large number of dynamic and often interactive group members. The nature of these applications is such that participants are not interested in all the content transmitted. We examine two currently available techniques to scope delivery of content to interested receivers in IP multicast: filtering, where data is filtered by middleware before passed to the application, and addressing, where data is routed only to those receivers that express their interest. We propose a framework that models large-scale application behavior. We use this framework to evaluate the performance of these applications and related protocols when the network is capable of filtering or addressing. Our results show that the current Internet architecture does not efficiently support large-scale applications because it can not efficiently manage multiple multicast groups. We show that addressing is preferred to filtering given that groups are easy to create and manage. We highlight future areas of research to bring about this change.

This paper proposes a novel approach centered on multimedia internetworking for the development of distributed virtual instruments. Multimedia internetworking refers to network infrastructures, protocols, models, applications and techniques being currently deployed over the Internet to support multimedia applications, e.g., videoconferencing, video-on-demand, shared workspaces. It is applied to broaden the concept of virtual instrument and enable new measurement patterns leveraging efficiency and interactivity. A Distributed Virtual Instrument (DVI) is a virtual instrument split into possibly multiple and independent parts, sender and receiver, which are linked by real-time continuous media and control streams. Senders and receivers are built by using open, composable and modular components based on a time sensitive actor framework and glued by multimedia middleware. A prototype is described to demonstrate the potential and the benefits of the proposed approach.

In 1993 the GARR (Italian Academic and Research Network) was connected to MBONE (worldwide IP multicast enabled network backbone). Since then, and especially in the last few years, this network has been used to test and develop new IP multicast routing protocols and applications. However it was a tunnel-based solution, not suitable for taking advantage of the new potential of IP multicast nor for use by a large community. This paper describes the work done for a migration to a native IP multicast routing deployment, highlighting the solutions adopted when dealing w ith the implementation problems and the complex wide area network management. Aim of this work is also to ensure a native IP multicast connection with the other Research Networks and the rest of Internet.

Multicast potentially optimises bandwidth consumption and node resources, when several users simultaneously participate in a communication session. Nevertheless, contrary to the expectations, IP multicast has not experienced widespread deployment, with the exception of IPTV. On the other hand, emerging Wireless Sensor Network (WSN) applications could greatly benefit from multicast and constitute another field where multicast can be an effective and efficient technique. The questions are: do multicast advantages hold in WSN scenarios? Can we use IP Multicast functionality in WSNs? This paper discusses and evaluates the use of multicast in WSNs.

The paper gives an introduction and overview of the activities conducted under the PIPVIC (Piloting IP Videoconferencing) project. The partner sites - who represent a range of UK Higher Education Institutions - are using IP multicast conferencing tools to support a range of collaborative teaching, research and administrative activities over a period of 5 months. The aim of the project i s to understand the potential issues involved in running a large-scale IP videoconferencing service, and to document working local set-up and support requirements for sites using such a service. The JANET Videoconferencing Strategy (1), as approved b y the ACN and JISC, recognises that, given the projected improvements in the packaging of IP-based v ideoconferencing software, there is a need to undertake piloting activities to understand potential i ssues involved in running a large scale IP videoconferencing service. The PIPVIC project was s et up to address this need, and to: • Pilot and assess IP ...

Overlay networks among cooperating hosts have recently emerged as a viable solution to several challenging problems, including multicasting, routing, content distribution, and peer-topeer services. Application-level overlays, however, incur a performance penalty over router-level solutions. This paper quantifies and explains this performance penalty for overlay multicast trees via: 1) Internet experimental data; 2) simulations; and 3) theoretical models. We compare a number of overlay multicast protocols with respect to overlay tree structure, and underlying network characteristics. Experimental data and simulations illustrate that the mean number of hops and mean per-hop delay between parent and child hosts in overlay trees generally decrease as the level of the host in the overlay tree increases. Overlay multicast routing strategies, overlay host distribution, and Internet topology characteristics are identified as three primary causes of the observed phenomenon. We show that this phenomenon yields overlay tree cost savings: Our results reveal that the normalized cost ( ) ( ) is 0 9 for small , where ( ) is the total number of hops in all overlay links, ( ) is the average number of hops on the source to receiver unicast paths, and is the number of members in the overlay multicast session. This can be compared to an IP multicast cost proportional to 0 6 to 0 8 . Index Terms-Economies of scale, group communication, Internet multicast, overlay multicast, overlay networks.