Dynamic batching policies for an on-demand video server

In a video-on-demand environment, continuous delivery of video streams to the clients is guaranteed by sufficient reserved network and server resources. This leads to a hard limit on the number of streams that a video server can deliver. Multiple client requests for the same video can be served with a single disk I/O stream by sending (multicasting) the same data blocks to multiple clients (with the multicast facility, if present in the system). This is achieved by batching (grouping) requests for the same video that arrive within a short time. We explore the role of customer-waiting time and reneging behavior in selecting the video to be multicast. We show that a first come, first served (FCFS) policy that schedules the video with the longest outstanding request can perform better than the maximum queue length (MQL) policy that chooses the video with the maximum number of outstanding requests. Additionally, multicasting is better exploited by scheduling playback of the most popular videos at predetermined, regular intervals (hence, termed FCFS-). If user reneging can be reduced by guaranteeing that a maximum waiting time will not be exceeded, then performance of FCFS- is further improved by selecting the regular playback intervals as this maximum waiting time. For an empirical workload, we demonstrate a substantial reduction (of the order of 60%) in the required server capacity by batching.     

Issues in the design of a storage server for video-on-demand

We examine issues related to the design of a storage server for video-on-demand (VOD) applications. The storage medium considered is magnetic disks or arrays of disks. We investigate disk scheduling policies, buffer management policies and I/O bus protocol issues. We derive the number of sessions that can be supported from a single disk or an array of disks and determine the amount of buffering required to support a given number of users. Furthermore, we propose a scheduling mechanism for disk accesses that significantly lowers the buffer-size requirements in the case of disk arrays. The buffer size required under the proposed scheme is independent of the number of disks in the array. This property allows for striping video content over a large number of disks to achieve higher concurrency in access to a particular video object. This enables the server to satisfy hundreds of independent requests to the same video object or to hundreds of different objects while storing only one copy of each video object. The reliability implications of striping content over a large number of disks are addressed and two solutions are proposed. Finally, we examine various policies for dealing with disk thermal calibration and the placement of videos on disks and disk arrays.     

Metropolitan area video-on-demand service using pyramid broadcasting

Pyramid broadcasting is a new way of giving video-on-demand service on a metropolitan scale. We multiplex the most frequently requested movies on the network, gaining a radical improvement in access time and bandwidth use. This is achieved by using storage at the receiving end. As the available bandwidth increases, the improvement in access time is exponential instead of linear as in conventional broadcasting. The larger the bandwidth of the network is, the better gain in the access time due to pyramid broadcasting. As the access-time requirement decreases, the bandwidth in conventional broadcasting increases linearly, while the bandwidth in pyramid broadcasting increases only logarithmically. We provide analytical and experimental evaluations of pyramid broadcasting based on its implementation on an Ethernet LAN.     

Design and analysis of a video-on-demand server

The availability of high-speed networks, fast computers and improved storage technology is stimulating interest in the development of video on-demand services that provide facilities similar to a video cassette player (VCP). In this paper, we present a design of a video-on-demand (VOD) server, capable of supporting a large number of video requests with complete functionality of a remote control (as used in VCPs), for each request. In the proposed design, we have used an interleaved storage method with constrained allocation of video and audio blocks on the disk to provide continuous retrieval. Our storage scheme interleaves a movie with itself (while satisfying the constraints on video and audio block allocation. This approach minimizes the starting delay and the buffer requirement at the user end, while ensuring a jitter-free display for every request. In order to minimize the starting delay and to support more non-concurrent requests, we have proposed the use of multiple disks for the same movie. Since a disk needs to hold only one movie, an array of inexpensive disks can be used, which reduces the overall cost of the proposed system. A scheme supported by our disk storage method to provide all the functions of a remote control such as “fast-forwarding”, “rewinding” (with play “on” or “off”), “pause” and “play” has also been discussed. This scheme handles a user request independent of others and satisfies it without degrading the quality of service to other users. The server design presented in this paper achieves the multiple goals of high disk utilization, global buffer optimization, cost-effectiveness and high-quality service to the users.     

Efficient video file allocation schemes for video-on-demand services

A video-on-demand (VOD) server needs to store hundreds of movie titles and to support thousands of concurrent accesses. This, technically and economically, imposes a great challenge on the design of the disk storage subsystem of a VOD server. Due to different demands for different movie titles, the numbers of concurrent accesses to each movie can differ a lot. We define access profile as the number of concurrent accesses to each movie title that should be supported by a VOD server. The access profile is derived based on the popularity of each movie title and thus serves as a major design goal for the disk storage subsystem. Since some popular (hot) movie titles may be concurrently accessed by hundreds of users and a current high-end magnetic disk array (disk) can only support tens of concurrent accesses, it is necessary to replicate and/or stripe the hot movie files over multiple disk arrays. The consequence of replication and striping of hot movie titles is the potential increase on the required number of disk arrays. Therefore, how to replicate, stripe, and place the movie files over a minimum number of magnetic disk arrays such that a given access profile can be supported is an important problem. In this paper, we formulate the problem of the video file allocation over disk arrays, demonstrate that it is a NP-hard problem, and present some heuristic algorithms to find the near-optimal solutions. The result of this study can be applied to the design of the storage subsystem of a VOD server to economically minimize the cost or to maximize the utilization of disk arrays.     

On the design of a low-cost video-on-demand storage system

Recent advances in storage technology, coupled with the dramatic increase in the bandwidth of networks, now make it possible to provide "video-on-demand" service to viewers. A video-on-demand server is a computer system that stores videos in compressed digital form and provides support for various portions of compressed video data to be accessed and transmitted concurrently. We present a low-cost storage architecture for a video-on-demand server that relies principally on disks. The high bandwidths of disks in conjunction with a clever strategy for striping videos on them enables simultaneous access and transmission of portions of a video, separated by fixed time intervals. We also present schemes for implementing VCR-like functions including fast forward, rewind, and pause, and extend our schemes to the case in which videos have different rate requirements.     

Adaptive piggybacking: a novel technique for data sharing in video-on-demand storage servers

Recent technology advances have made multimedia on-demand services, such as home entertainment and home-shopping, important to the consumer market. One of the most challenging aspects of this type of service is providing access either instantaneously or within a small and reasonable latency upon request. We consider improvements in the performance of multimedia storage servers through data sharing between requests for popular objects, assuming that the I/O bandwidth is the critical resource in the system. We discuss a novel approach to data sharing, termed adaptive piggybacking, which can be used to reduce the aggregate I/O demand on the multimedia storage server and thus reduce latency for servicing new requests.     

Design and analysis of a look-ahead scheduling scheme to support pause-resume for video-on-demand applications

In a video-on-demand (VOD) system, subscribers can choose both the movie they wish to view and the time they wish to view it. In such an environment there are invariably hot videos which are requested by many viewers. The requirement that each viewer be able to independently pause the video at any instant and later resume the viewing with little delay can cause difficulties in batching viewers for each showing. Under batching, a single video stream is shared by multiple concurrent viewers and a resume request has to wait for additional stream capacity to become available before actual resumption can occur. The conventional approach to the support of on-demand pause-resume provides one video access stream to disks for each video request. This can greatly increase the disk arm requirements of a VOD system. In this paper, we propose a more efficient mechanism to support the pause-resume feature usinglook-ahead scheduling withlook-aside buffering. The idea is to use buffering to increase the number of concurrent viewers supportable. The concept of look-ahead scheduling is not to back up each viewer with a real stream capacity so he can pause and resume at any time, but rather with a (look-ahead) stream that is currently being used for another showing which is close to completion. Before the look-ahead stream becomes available, the pause and resume features have to be supported by the original stream through (look-aside) buffering of the missed content. It is shown via simulations that the proposed scheme can provide a substantially greater throughput than the approach without batching. Furthermore, for a given amount of buffer, the improvement in throughput grows more than linearly with the stream capacity of the server. In other words, the look-ahead stream scheduling scheme operates with good economy of scale because it is easier to form look-ahead streams for video servers with larger stream capacity.     

A circular skip-cluster scheme to support video-on-demand services

The major drawback of the existing cluster placement scheme is the long response time caused by admission control if the number of clusters and the number of users are large. A circular skip-cluster placement scheme is proposed to reduce the size of the data buffer as well as the system response time. Furthermore, the popularity of each video is different in the real world. We propose a new popularity-based data allocation scheme to allocate data units within a cluster such that the corresponding data units of these popular videos are stored in those cylinders at one end of each cluster. Due to a higher spatial locality within these hot cylinders, some data units requested by the users are stored in the same cylinder such that one seek operation, one rotation, and one transfer operation are required to retrieve these data units. Therefore, the time required to retrieve data for these requests can be reduced, thus also reducing the system response time. Based on our results, the buffer size and the system response time can be reduced by half or more. These findings are essential for constructing video-on-demand systems that provide satisfactory performance.     

Impact of video scheduling on bandwidth allocation for multiplexed MPEG streams

We present efficient schemes for scheduling the delivery of variable-bit-rate MPEG-compressed video with stringent quality-of-service (QoS) requirements. Video scheduling is being used to improve bandwidth allocation at a video server that uses statistical multiplexing to aggregate video streams prior to transporting them over a network. A video stream is modeled using a traffic envelope that provides a deterministic time-varying bound on the bit rate. Because of the periodicity in which frame types in an MPEG stream are typically generated, a simple traffic envelope can be constructed using only five parameters. Using the traffic-envelope model, we show that video sources can be statistically multiplexed with an effective bandwidth that is often less than the source peak rate. Bandwidth gain is achieved without sacrificing the stringency of the requested QoS. The effective bandwidth depends on the arrangement of the multiplexed streams, which is a measure of the lag between the GOP periods of various streams. For homogeneous streams, we give an optimal scheduling scheme for video sources at a video-on-demand server that results in the minimum effective bandwidth. For heterogeneous sources, a sub-optimal scheduling scheme is given, which achieves acceptable bandwidth gain. Numerical examples based on traces of MPEG-coded movies are used to demonstrate the effectiveness of our schemes.     

Efficient support for interactive scanning operations in MPEG-based video-on-demand systems

In this paper, we present an efficient approach for supporting fast-scanning (FS) operations in MPEG-based video-on-demand (VOD) systems. This approach is based on storing multiple, differently encoded versions of the same movie at the server. A normal version is used for normal playback, while several scan versions are used for FS. Each scan version supports forward and backward FS at a given speedup. The server responds to an FS request by switching from the normal version to an appropriate scan version. Scanning versions are produced by encoding a sample of the raw frames using the same GOP pattern of the normal version. When a scanning version is decoded and played back at the normal frame rate, it gives a perceptual motion speedup. By being able to control the traffic envelopes of the scan versions, our approach can be integrated into a previously proposed framework for distributing archived, MPEG-coded video streams. FS operations are supported using no or little extra network bandwidth beyond what is already allocated for normal playback. Mechanisms for controlling the traffic envelopes of the scan versions are presented. The actions taken by the server and the client's decoder in response to various types of interactive requests are described in detail. The latency incurred in implementing various interactive requests is shown to be within an acceptable range. Striping and disk-scheduling strategies for storing various versions at the server are presented. Issues related to the implementation of our approach are discussed.     

An effective admission control mechanism for variable-bit-rate video streams

For admission control in real-time multimedia systems, buffer space, disk bandwidth and network bandwidth must be considered. The CBR-based mechanisms do not use system resources effectively, since media data is usually encoded with VBR compression techniques. We propose an admission control mechanism based on a VBR data model that has a dynamic period length. In our mechanism, the period can be adaptively changed to maximize the performance, considering both disk bandwidth and buffer space. To compare the performance, extensive simulations are conducted on RR, SCAN, and GSS schemes which have the dynamic period length and the static period length.     

Spatio-temporal effects of multimedia objects storage and delivery for video-on-demand systems

As the number of video streams to be supported by a digital video delivery system (DVDS) increases, an improved understanding of the necessity for reliable and cost-efficient support for a considerable number of video streams (in the magnitude of tens of thousands), and the dependency largely on software capabilities emerges. Even in the presence of an optimal hardware configuration, or model, and associated costs, using software to exploit the underlying hardware capabilities is of paramount importance. Although a number of DVDSs have become operational, their ability to deliver the required services mainly depends on the small number of streams supported and the hardware trade-offs. It is imperative that current software developments account for the eventual scalability of the number of video streams without commensurate increase in hardware. In this paper, we present strategies for the management of video streams in order to maintain and satisfy their space and time requirements. We use a DVDS architectural model with functionally dichotomized nodes: a single-node partition is responsible for data retrieval, while the remaining partition of nodes accepts user requests, determines object locations, and routes requests through the network that connects both partitions. We present a detailed analysis of the issues related to queuing I/O requests and data buffering. The discussion includes the requirements for arranging and scheduling I/O requests and data buffers, with the objective of guaranteeing the required data availability rates for continuous media display.     

Fast techniques for the optimal smoothing of stored video

Work-ahead smoothing is a technique whereby a server, transmitting stored compressed video to a client, utilizes client buffer space to reduce the rate variability of the transmitted stream. The technique requires the server to compute a schedule of transfer under the constraints that the client buffer neither overflows nor underflows. Recent work established an optimal off-line algorithm (which minimizes peak, variance and rate variability of the transmitted stream) under the assumptions of fixed client buffer size, known worst case network jitter, and strict playback of the client video. In this paper, we examine the practical considerations of heterogeneous and dynamically variable client buffer sizes, variable worst case network jitter estimates, and client interactivity. These conditions require on-line computation of the optimal transfer schedule. We focus on techniques for reducing on-line computation time. Specifically, (i) we present an algorithm for precomputing and storing the optimal schedules for all possible client buffer sizes in a compact manner; (ii) we show that it is theoretically possible to precompute and store compactly the optimal schedules for all possible estimates of worst case network jitter; (iii) in the context of playback resumption after client interactivity, we show convergence of the recomputed schedule with the original schedule, implying greatly reduced on-line computation time; and (iv) we propose and empirically evaluate an “approximation scheme” that produces a schedule close to optimal but takes much less computation time.     

Dynamic playout scheduling algorithms for continuous multimedia streams

In this paper, we investigate a playout scheduling framework for supporting the continuous and synchronized presentations of multimedia streams in a distributed multimedia presentation system. We assume a situation in which the server and network transmissions provide sufficient support for the delivery of media objects. In this context, major issues regarding the enforcement of the smooth presentation of multimedia streams at client sites must be addressed to deal with rate variance of stream presentations and delay variance of networks. We develop various playout-scheduling algorithms that are adaptable to quality-of-service parameters. The proposed algorithms permit the local adjustment of unsynchronized presentations by gradually accelerating or retarding presentation components, rather than abruptly skipping or pausing the presentation materials. A comprehensive experimental analysis of the proposed algorithms demonstrates that our algorithms can effectively avoid playout gaps (or hiccups) in the presentations. This scheduling framework can be readily used to support customized multimedia presentations.     

Design and analysis of permutation-based pyramid broadcasting

Periodic broadcasting can be used to support near-video-on-demand for popular videos. For a given bandwidth allocation, pyramid broadcasting schemes substantially reduce the viewer latency (waiting) time compared to conventional broadcasting schemes. Nevertheless, such pyramid schemes typically have substantial storage requirements at the client end, and this results in set-top boxes needing disks with high transfer rate capabilities. In this paper, we present a permutation-based pyramid scheme in which the storage requirements and disk transfer rates are greatly reduced, and yet the viewer latency is also smaller. Under the proposed approach, each video is partitioned into contiguous segments of geometrically increasing sizes, and each segment is further divided into blocks, where a block is the basic unit of transmission. As in the original pyramid scheme, frequencies of transmission for the different segments of a video vary in a manner inversely proportional to their size. Instead of transmitting the blocks in each segment in sequential order, the proposed scheme transmits these blocks in a prespecified cyclic permutation to save on storage requirements in the client end. Performance analyses are provided to quantify the benefits of the new scheme.     

On periodic resource scheduling for continuous-media databases

The Enhanced Pay-Per-View (EPPV) model for providing continuous-media services associates with each continuous-media clip a display frequency that depends on the clip's popularity. The aim is to increase the number of clients that can be serviced concurrently beyond the capacity limitations of available resources, while guaranteeing a constraint on the response time. This is achieved by sharing periodic continuous-media streams among multiple clients. The EPPV model offers a number of advantages over other data-sharing schemes (e.g., batching), which make it more attractive to large-scale service providers. In this paper, we provide a comprehensive study of the resource-scheduling problems associated with supporting EPPV for continuous-media clips with (possibly) different display rates, frequencies, and lengths. Our main objective is to maximize the amount of disk bandwidth that is effectively scheduled under the given data layout and storage constraints. Our formulation gives rise to -hard combinatorial optimization problems that fall within the realm of hard real-time scheduling theory. Given the intractability of the problems, we propose novel heuristic solutions with polynomial-time complexity. We also present preliminary experimental results for the average case behavior of the proposed scheduling schemes and examine how they compare to each other under different workloads. A major contribution of our work is the introduction of a robust scheduling framework that, we believe, can provide solutions for a variety of realistic EPPV resource-scheduling scenarios, as well as any scheduling problem involving regular, periodic use of a shared resource. Based on this framework, we propose various interesting research directions for extending the results presented in this paper. Received June 9, 1998 / Accepted October 13, 1998     

Using rotational mirrored declustering for replica placement in a disk-array-based video server

In a video-on-demand (VOD) environment, disk arrays are often used to support the disk bandwidth requirement. This can pose serious problems on available disk bandwidth upon disk failure. In this paper, we explore the approach of replicating frequently accessed movies to provide high data bandwidth and fault tolerance required in a disk-array-based video server. An isochronous continuous video stream imposes different requirements from a random access pattern on databases or files. Explicitly, we propose a new replica placement method, called rotational mirrored declustering (RMD), to support high data availability for disk arrays in a VOD environment. In essence, RMD is similar to the conventional mirrored declustering in that replicas are stored in different disk arrays. However, it is different from the latter in that the replica placements in different disk arrays under RMD are properly rotated. Combining the merits of prior chained and mirrored declustering methods, RMD is particularly suitable for storing multiple movie copies to support VOD applications. To assess the performance of RMD, we conduct a series of experiments by emulating the storage and delivery of movies in a VOD system. Our results show that RMD consistently outperforms the conventional methods in terms of load-balancing and fault-tolerance capability after disk failure, and is deemed a viable approach to supporting replica placement in a disk-array-based video server.     

An optimal bandwidth allocation strategy for the delivery of compressed prerecorded video

The transportation of prerecorded, compressed video data without loss of picture quality requires the network and video servers to support large fluctuations in bandwidth requirements. Fully utilizing a client-side buffer for smoothing bandwidth requirements can limit the fluctuations in bandwidth required from the underlying network and the video-on-demand servers. This paper shows that, for a fixed-size buffer constraint, the critical bandwidth allocation technique results in plans for continuous playback of stored video that have (1) the minimum number of bandwidth increases, (2) the smallest peak bandwidth requirements, and (3) the largest minimum bandwidth requirements. In addition, this paper introduces an optimal bandwidth allocation algorithm which, in addition to the three critical bandwidth allocation properties, minimizes the total number of bandwidth changes necessary for continuous playback. A comparison between the optimal bandwidth allocation algorithm and other critical bandwidth-based algorithms using 17 full-length movie videos and 3 seminar videos is also presented.     

A buffer-inventory-based dynamic scheduling algorithm for multimedia-on-demand servers

We present a producer-consumer model of multimedia-on-demand (MOD) servers. The producer retrieves media data from a disk and places it into a set of buffers, while the consumer sends out the data in the buffers to the users. We develop for the producer a buffer-inventory-based dynamic scheduling (BIDS) algorithm that guarantees non-zero inventory and non-overflow of data in the buffers to meet the continuity requirement and no-loss of data for each media stream. The algorithm can deal with heterogeneous me dia streams as well as the transient circumstances upon service completions and arrivals of new requests. To smooth out the impact of bursty data of variable-bit-rate media streams and therefore increase the maximum admissible load of requests, we also introduce into the scheduling scheme a time-scale-dependent peak consumption rate and a virtual cycle time. Based on BIDS, an effective admission control mechanism can be easily established by checking two simple conditions respectively on the overall system load and buffer size. Our algorithm is very easy to implement. Experiments carried out with an actual disk system and real video stream data verify that it is more robust compared to static scheduling algorithms previously proposed in the literature, especially when handling variable-bit-rate media streams.