This paper proposes a method to reduce the playback suspension in a Video-on-Demand system based on the Peer-to-Peer technology (P2P VoD). Our main contribution is twofold. The first is the proposal of a hierarchical P2P architecture with the notion of dynamic swarms. Swarm is a group of peers to have similar playback position and those swarms are connected with an overlay so that requested pieces are forwarded from a swarm to another swarm in a bucket brigade manner, where the forward of pieces is regulated by the super-peer (SP) of each swarm. The second contribution is the proposal of a match making scheme between requests and uploaders. The simulation result indicates that the proposed scheme reduces the total waiting time of a randomized scheme by 24% and the load of the media server by 76%.

In this paper, we propose a new buffer map notification scheme for Peer-to-Peer Video-on-Demand systems (P2P VoDs) which support VCR operations such as fast-forward, fast-backward, and seek. To enhance the fluidity of such VCR operations, we need to refine the size of each piece as small as possible. However, such a refinement significantly degrades the performance of buffer map notification schemes with respect to the overhead, piece availability and the efficiency of resource utilizations. The basic idea behind our proposed scheme is to use a piece-based buffer map with a segment-based buffer map in a complementary manner. The result of simulations indicates that the proposed scheme certainly increases the accuracy of the information on the piece availability in the neighborhood with a sufficiently low cost, which reduces the intermittent waiting time of each peer by more than 40% even under a situation in which 50% of peers conduct the fast-forward operation over a range of 30% of the entire video.

In this paper, we consider Peer-to-Peer Video-on-Demand (P2P VoD) systems based on the BitTorrent file sharing protocol. Since the Rarest First policy adopted in the original BitTorrent protocol frequently fails to collect pieces corresponding to a video file by their playback time, we need to develop a new piece selection rule particularly designed for P2P VoDs. In the proposed scheme, we assume the existence of a media server which can upload any piece upon request, and try to bound the load of such media server with two techniques. The first technique is to estimate pieces which are not held by any peer and prefetch them from the media server. The second technique is to switch the mode of each peer according to the estimated size of the P2P network. The performance of the proposed scheme is evaluated by simulation.

In designing a video-on-demand system, one of the major challenges is how to reduce the client's waiting time maintaining the concurrently used channels. For this reason, the hybrid architectures which integrate the multicast streams with the unicast streams were suggested in order to improve channel efficiency in recent years. In combining multicast with unicast, the ways to group the channels together are important so that more clients can share the multicast transmission channels. This paper proposes a hybrid video-on-demand system which gathers the unicast and multicast transmission channels efficiently by using dynamic channel allocation architecture. The newly proposed architecture can reduce the average client's waiting time significantly. The numerical results demonstrate that the dynamic channel allocation architecture in some case (e.g., 100-channel and 10-video system at 0.5 requests/second) achieves performance gain of 551% compared to existing architecture. This paper presents procedure of channel release and reuse, performance analysis, and simulation results of the dynamic channel allocation architecture.

This paper proposes a scalable video delivery scheme, named P2PVD, for large-scale video-on-demand applications based on the emerging peer-to-peer structure and characteristic user behaviors. Two types of orders are permitted in P2PVD: reserved and urgent. Reserved orders are encouraged with a lower price policy, which smoothes the network traffic and reduces the server load. The requesting peers use delay-aware dynamic parallel transmission to serve the reserved and urgent orders, and supplying peers employ three priority rules to increase the capacity of P2PVD. Experimental results indicate that P2PVD is scalable.

In providing video programs to a number of clients through networks, a broadcasting approach is more appropriate than a true-video-on-demand approach in efficient use of bandwidth. However, the broadcasting approach also needs excessively wide bandwidth if many video programs are to be transmitted. This study presents a very simple but novel architecture called dynamic channel broadcasting, for video-on-demand systems. The proposed architecture uses both static and dynamic broadcasting channels to improve the efficiency of channels. The proposed architecture eliminates the necessity of dividing each video into segments and switching channels frequently unlike the pyramid broadcasting, the skyscraper broadcasting and the harmonic broadcasting. Also this new architecture needs a smaller buffer size. The numerical results demonstrate that the newly proposed approach in some cases requires only 14% of the bandwidth required for the conventional broadcasting while maintaining the start-up latency.

In this paper, we study an effective video-on-demand traffic control algorithm using the metadata over the network supporting bandwidth-renegotiations. The proposed algorithm includes bandwidth smoothing, bandwidth-burstiness estimation and rate adaptation algorithms. The proposed video-on-demand server has not only video database but also metadata database that includes coding information of the compressed video in video database and the traffic burst characteristics with respect to control parameters of the bandwidth smoothing algorithm. Thus, we can predict the traffic properties accurately with a low computational complexity by using the stored metadata, and then determine the efficient bandwidth renegotiating variables such as the renegotiating instants and the required bandwidth in terms of network utilization and video-on-demand service quality. In addition, we present a rate adaptation algorithm that pursues an effective trade-off between spatial and temporal qualities of the decoded video to improve the perceptual video quality when the bandwidth request is rejected.

This paper presents a novel data placement and retrieval scheme for video-on-demand systems. In particular, a zone-based data placement scheme is employed to reduce the average seek time of the disk array storage system and thus increase the disk access bandwidth to allow the video server provide more services of video programs concurrently. Furthermore, due to the inherent nature of video access, a popular program always requires more accesses and therefore occupies more disk I/O bandwidth as request for serving such program increases. A new retrieval strategy is proposed to maintain a single copy of each video program disregarding the popularity of the video programs, and to achieve maximum I/O throughput of the video server.

To understand the characteristics of a multimedia service, such as the large volume of data transfer and real-time constraints, it is necessary to have a performance evaluation tool for an HDD. Our HDD simulator is running on a PC operated on FreeBSD UNIX OS. We first investigate the seek time and the sustained rate of HDDs and then evaluate the performance of an HDD for an experimental VOD system. Applying the experimental results, we find the bottleneck of an HDD, and then suggest what HDDs are to be selected for a VOD system.

B-ISDN is expected to be applied in the near future to video delivery systems for the broadcast of news and television programs. The demand for such services is increasing, and in particular, on-demand services are becoming more desirable. On-demand services allow viewers to request their favorite programs at the time that is convenient, hence catering for the wide range of modern lifestyles. As for on-demand services, there already exist Video on Demand (VoD) systems such as the original VoD or Near VoD. However, such systems have not yet been widely implemented because of the inefficient cost of communication resources, and storage. The authors' research is aimed at producing an efficient VoD system based on a high speed network. We are focused in particular on the forms of data transmission, and in this paper, we propose a new VoD system called Burst VoD. Burst VoD aggressively utilizes the multicasting technique, and involves dividing the program resource data into block files and transmitting them to viewer terminals as burst traffic over a high speed network. Simulation results comparing Burst VoD with conventional VoD show that Burst VoD achieves lower request blocking rates, efficient utilization of networks with multicasting, and almost on-demand response time to requests.

We present an architecture of a VOD system employing proxy servers. The proposed VOD system provides efficient and reliable VOD services and solves the problems caused by traditional VOD systems of centralized, hierarchical or distributed architecture. The proxy servers are placed between video servers and user systems. The proxy server is a small size video server that has not only caching function but also intelligence such as VCR-like video stream control or navigation of other proxy/video servers to search for a selected video program. Using a VOD system of the proposed architecture, the VOD services can be provided to more users because it reduces the workload of video servers and network traffic. We provide the performance model of the system. Service availability is also analyzed. The proposed architecture shows better performance and availability than the traditional VOD architectures.

This paper investigates the lossless video aggregation and transport, a concept of transmitting a group of video sessions as a bundle over a communication network. We focus on the design of optimal transmission plan with minimum receiver buffer requirement for stored variable bit-rate (VBR) videos over a constant bit-rate (CBR) channel without incurring buffer underflow and overflow. For a single video, an efficient algorithm is proposed to calculate the optimal transmission plan in only O(N) time, a significant improvement of the previous result with time complexity of O(N2log N). For multiple video sessions, we propose an aggregation scheme to calculate the changes of optimal transmission plans at the joints or separations, where a session bundled in the aggregated video stream begins or stops. The experimental results show that our algorithms stand out in terms of simplicity and efficiency.

Small-scale video on demand system will be necessary in the future. Cluster drives, which use optical disk drives, are a good buffer memory for this purpose because the cost per megabyte is low. An ordinary optical cluster drive has many SCSI buses and up to seven optical drives are connected to each SCSI bus. One drive from each bus is assembled to make a group of a cluster drive. The difference betweeen SCSI bus data transfer rate and sustained disk transfer rate enables the cluster drive to be simplified. Several drives on an SCSI bus make a sub-group. The video data is striped onto those sub-groups. When the total data transfer rate from disks within a sub-group exceeds the bus transfer rate, some drives can not acquire the bus. When drives connected to one SCSI bus are not identical, the block size of the data to be recorded on each drive has to be adjusted so that the maximum effective data transfer rate can be obtained. When the cycle times of a slow and fast drive are set identical, the effective data transfer rate is maximum, where one cycle consists of command time, minimum bus free time, disk read time, and bus transfer time.

In this paper, we point out an architecture optimization problem for networks delivering services such as Video-On-Demand or, more precisely, two intertwined problems, i.e., the storage allocation of the videos among the storage nodes of the network and the choice of the network topology. We present and investigate the properties of a genetic algorithm which can handle such problems. This algorithm, as well as a greedy heuristics and simulated annealing, are then used to derive solutions in function of link and node cost parameters in a 36-node network. The results show that genetic algorithms are an effective class of algorithms for such problems, and possibly many other topology optimization problems.

This paper presents the system issues involved with the transmission of pre-encoded VBR MPEG video using CBR service. Conventional wisdom suggests that lossless delivery of VBR video using CBR service requires bandwidth to be reserved at the peak rate resulting in low bandwidth utilization. We calculate the minimum rate at which bandwidth must be reserved on a network in order to provide continuous playback of an MPEG encoded video bitstream. Simulation results using the frame size traces from several pre-encoded MPEG bitstreams and several buffer sizes demonstrate that this minimum reservation rateis much lower than the peak rate when a relatively small playback buffer size is used, resulting in much higher bandwidth utilization. Procedures for performing connection setup and lossless realtime video playback between the video server and the client are outlined. Methods for incorporating VCR-like features such as pauseandfast forward/reversefor Video-on-Demand (VoD) applications are presented.

This paper illustrates activities and accomplishments being made by DAVIC, a non-profit organization pushing forward its open, international, cross-industry standards for audio-visual information systems, of which video-on-demand is the representative. Core technologies selected in its firstly published specifications and their interoperability aspects are summarized here. Preliminary results in our interoperability testing are also shown. Finally, we touch upon the coming work plan of DAVIC which covers wider range of access network capabilities and service domains, e.g. internet.

This paper studies a video storage problem that occurs in Video-on-Demand (VOD) networks and in other distributed database systems. Videos should be stored in order to respect various constraints, especially available storage and transmission capacities. We show there exists an algorithm to solve this combinatorial problem through a pricing mechanism and that it converges to a solution under some general conditions. Simulation results with up to 43-node networks and up to 300 videos show that the algorithm is fast.