Staggered push - A linearly scalable architecture for push-based parallel video servers

Dept. of Inf. Eng., Chinese Univ. of Hong Kong, Shatin, China
IEEE Transactions on Multimedia (Impact Factor: 2.3). 03/2003; 4(4):423 - 433. DOI: 10.1109/TMM.2002.806533
Source: IEEE Xplore


With the rapid performance improvements in low-cost PCs, it becomes increasingly practical and cost-effective to implement large-scale video-on-demand (VoD) systems around parallel PC servers. This paper proposes a novel parallel video server architecture where video data are striped across an array of autonomous servers connected by an interconnection network. To coordinate data transmissions from multiple autonomous servers to a client station, a staggered push scheduling algorithm is proposed. A system model is constructed to quantify the performance of the architecture. Unlike most studies, this work does not assume the existence of a global clock among the servers and tackles two problems arising from server asynchrony: inconsistent schedule assignment and traffic overlapping. The former problem is solved by using an admission scheduler and the latter problem is solved by an over-rate transmission scheme. Analytical results prove a remarkable property of the staggered push architecture: as long as the network has sufficient capacity, the system can be scaled up linearly to an arbitrary number of servers. Design examples and numerical results are used to evaluate the proposed architecture under realistic assumptions and to compare it against other architecture.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Real-time multimedia transport has stringent quality of service requirements, which are generally not supported by current network architectures. In emerging mobile ad hoc networks, frequent topology changes and link failures cause severe packet losses, which degrade the quality of received media. However, in such mesh networks, there usually exist multiple paths between any source and destination nodes. Such path diversity has been demonstrated to be effective in combating congestion and link failures for improved media quality. In this paper, we present a new protocol to facilitate multipath transport of real-time multimedia data. The proposed protocol, the multiflow real-time transport protocol (MRTP), provides a convenient vehicle for real-time applications to partition and transmit data using multiple flows. We demonstrate through analysis that data partitioning, which is an essential function of MRTP, can effectively reduce the short-range dependence of multimedia data, thus improving its queueing performance in underlying networks. Furthermore, we show that a few flows are sufficient for MRTP to exploit most of the benefits of multipath transport. Finally, we present a comprehensive simulation study on the performance of MRTP under a mobile ad hoc network. We show that with one additional path, MRTP outperformed single-flow RTP by a significant margin.
    Preview · Article · May 2006 · IEEE Transactions on Multimedia
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Parallel video servers can achieve highly storage-saving and granularly load-balancing, but they suffer from a system expansion problem. As the number of users continuously increases, the system inevitably needs to expand the number of video servers. However, the expansion of a parallel video server system is not as simple as that of a replicated video server system. Hence, this work develops an efficient expansion algorithm, called the Cyclic Expansion Algorithm (CEA), for parallel video servers. The proposed CEA algorithm has several good features. First, the data layout of each video content exhibits periodicity. Consequently, the meta-data size of each video and the complexity of the CEA algorithm are reduced. Second, the number of required data movements during a system expansion is optimized. Third, the total number of required XOR recomputations for updating parity blocks during an expansion is also minimized. Additionally, the new CEA can be applied to a variety of distributed storage systems, such as the cloud-based storage systems using striping and parity check techniques.
    Full-text · Article · Jul 2012 · Journal of Parallel and Distributed Computing
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper examines the techniques for passing information from one web site to another, the methodologies that have been developed and the advances in technology that has occurred in relation to these techniques. This paper then utilises one of these technologies to develop an environment where the major techniques (Polling and Push) can be tested. This paper's research finds that Polling uses substantially more resources than Push, particularly in regard to server usage. In situations where this is an issue the use of Push can reduce server cost by at least a factor of 5 and possibly more in a real world application. The reduction in network cost is more difficult to determine using the tools in this paper but here the indications are that the reduction in cost may be by a factor of 3. Finally this paper presents the view that the techniques are complementary and both should be used within most real world application.
    No preview · Conference Paper · Jan 2013