Optimization Strategies for QoS-Sensitive VOD Service Delivery in Cooperative Environments

Jeon, Won J.

Permalink

https://hdl.handle.net/2142/11344 Copy

Description

Title

Optimization Strategies for QoS-Sensitive VOD Service Delivery in Cooperative Environments

Author(s)

Jeon, Won J.

Issue Date

2007-07

Keyword(s)

computer science

Abstract

Multimedia services such video-on-demand service to large number of clients require stable network bandwidth provision and short network delay. In order to achieve these goals, the object replication in overlay P2P networks in a good candidate solution for both service providers and customers. However, the characteristics of peers are different from traditional content distribution infrastructure, i.e., dynamic membership changes, asymmetric network bandwidth utilization, node heterogeneity, therefore dynamic change of service availability, etc., so that they make the multimedia content distribution more challenging. In this thesis, I present the collaborative object placement and replication schemes in collaborative computing environments with different scale of group size. In cooperative streaming and caching, we present a QoS-aware middleware for collaborative multimedia caching and caching service in heterogeneous computing environment, which not only reduces the initial delay of playing time, but also minimizes the delay jitter during playing time and net- work bandwidth utilization. With the publish/subscribe mechanism of cache service, we shows collaborative techniques for multimedia streaming, streaming scheduling and pre-fetching embedded in the QoS middleware architecture for dynamic group management. In distributed caching, we propose a novel distributed caching algorithm for multimedia streaming, supporting a large number of peers over P2P overlay networks. In order to facilitate multimedia streaming and downloading service from servers, our caching scheme determines the appropriate availability of cached stream segments in a cache community, determines the appropriate peers for cache replacement, and performs availability-aware cache replacement. It achieves a small access latency for stream retrieval, stable bandwidth provisioning during retrieval session, and load balancing of clients' requests among peers. In QoS-aware object replication, we formulate QoS-aware replication problem where each node specifies an upper bound on the time to access a given object. The problem is to minimize the number of replicas in order to satisfy these access deadlines. We show that this problem is intractable, since it is NP-complete. We present centralized as well as distributed solutions.

Type of Resource

text

Permalink

http://hdl.handle.net/2142/11344

Copyright and License Information

You are granted permission for the non-commercial reproduction, distribution, display, and performance of this technical report in any format, BUT this permission is only for a period of 45 (forty-five) days from the most recent time that you verified that this technical report is still available from the University of Illinois at Urbana-Champaign Computer Science Department under terms that include this permission. All other rights are reserved by the author(s).

Owning Collections