Principles of database buffer management
ACM Transactions on Database Systems (TODS)
Availability of a distributed computer system with failures
Acta Informatica
The coming revolution in interactive digital video
Communications of the ACM
Understanding fault-tolerant distributed systems
Communications of the ACM
Standards and the emergence of digital multimedia systems
Communications of the ACM - Special issue on digital multimedia systems
MPEG: a video compression standard for multimedia applications
Communications of the ACM - Special issue on digital multimedia systems
The design and analysis of an ATM multicast switch with adaptive traffic controller
IEEE/ACM Transactions on Networking (TON)
Operating system support for a video-on-demand file service
Multimedia Systems
Dynamic batching policies for an on-demand video server
Multimedia Systems
Fail-stop processors: an approach to designing fault-tolerant computing systems
ACM Transactions on Computer Systems (TOCS)
Byzantine generals in action: implementing fail-stop processors
ACM Transactions on Computer Systems (TOCS)
Multimedia Systems: An Overview
IEEE MultiMedia
Resource Allocation for Primary-Site Fault-Tolerant Systems
IEEE Transactions on Software Engineering
Buffer Management for Video Database Systems
ICDE '95 Proceedings of the Eleventh International Conference on Data Engineering
Buffer Management in Real-Time Databases
Buffer Management in Real-Time Databases
Hi-index | 0.00 |
Fault tolerance is an important design criterion for reliable and robust video-on-demand systems. Conventional fault-tolerant designs use eithera primary backup or an active replication method to provide system faulttolerance. However, these approaches suffer from low utilization of thebackup or replication system. In this paper we propose twoplayback-recovery schemes for distributed video-on-demand systemscalled the forward playback-recovery scheme and the backwardplayback-recovery scheme. Unlike conventional fault-tolerant designs,our schemes use existing playback resources to recover faulty playbackswithout allocating new resources, significantly reducing recovery overhead.To use the schemes effectively, we developed a distributed algorithmfor determining the order and gap information between the playbacks onthe distributed video-on-demand servers so that overhead for recoveringfrom a server failure can be minimized. This algorithmachieves N − 1 fault-tolerant resiliency for N-servervideo-on-demand systems. In addition, three server-recoverypolicies are also presented to guide surviving servers inapplying the proper scheme to recover faulty playbacks, thusreducing overall recovery costs. Simulation results show thatthe proposed recovery schemes are effective and useful indesigning fault-tolerant multiple-server video-on-demandsystems.