Browsing by Author "Dong, Yingfei"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item An Architecture for Proxy-Assisted Periodic Broadcast for Large Scale Video Streaming(2003-09-17) Kusmierek, Ewa; DuHung-Chang, David; Dong, YingfeiMany multimedia applications rely on video streaming techniques. However, large scale video delivery is still very challenging since it requires a large amount of resources such as storage space, network bandwidth and I/Obandwidth. In this paper we propose a proxy-assisted periodic broadcast architecture for video delivery to a large number of clients over the Internet. Our video delivery technique is based on a combination of periodic broadcast by central server and proxy server caching. A proxy server caches either part or the whole video based on the video popularity. We assume that each proxy server may have different capability and that the video popularity in each community can be different and dynamically changing. A video stored in the central server is partitioned into two parts, a server prefix and a server suffix, based onthe aggregated demand for the video from all communities. In principle, the server prefix is delivered by unicast and the server suffix is delivered by periodic broadcast. Such an approach allows to significantly reduce the required I/O bandwidth at a server. The combination of proxy prefix and server prefix defines a wide spectrum of different videodelivery modes. The transmission of a video can be either partially unicast or partially period broadcast depending on the relationship between proxy prefix and server prefix. We further define and solve the optimization problems for proxy prefix selection and server prefixselection in order to minimize the total resource requirements. Performance of our system is evaluated through a number of tests.Item An Efficient Client Collaboration Framework for Large-Scale Streaming(2005-08-08) Kusmierek, Ewa; Dong, Yingfei; DuHung-Chang, DavidIn this paper, we propose a loopback approach in a two-level streaming architecture to exploit collaborative client/proxy buffers for improving the quality and efficiency of large-scale streaming applications. At the upper level we use a Content Delivery Network (CDN) to deliver video from a central server to proxy servers. At the lower level a proxy server delivers video with the help of collaborative client caches. In particular, a proxy server and its clients in a local domain cache different portions of a video and form delivery loops. In each loop, a single video stream originates at the proxy, passes through a number of clients, and finally is passed back to the proxy. As a result, with limited bandwidth and storage space contributed by collaborative clients, we are able to significantly reduce the required network bandwidth, I/O bandwidth, and cache space of a proxy. Furthermore, we develop a local repair scheme to address the client failure issue for enhancing service quality and eliminating most required repairing load at the central server. For popular videos, our local repair scheme is able to handle most of single-client failures without service disruption and retransmissions from the central server. Our analysis and simulations have shown the effectiveness of the proposed Loopback Scheme.Item Full-Sharing Optimal Scheduling for VOD Service on Broadband Cable Networks(2002-06-27) Dong, Yingfei; Zhang, Zhi-Li; DuHung-Chang, DavidCable Access Networks (CANs) have become one of the most important forms of broadband services, which significantly improve the bandwidth to home and spur many multimedia applications to real life. %into reality. However, providing efficient Video-On-Demand (VOD) service on CANs isstill a challenging issue due to the lack of effective approaches to exploit the unique characteristics of CANs. Previous streaming approaches were intended for general network settings and do not consider these unique characteristics. In this study, we address the schedulingissue in the VOD service over CANs by fully utilizing the characteristics of CANs, such as subchannel sharing and potential large buffer space at CAN clients. We propose an optimal full-sharing scheduling approach that minimizes the bandwidth consumption of streaming sessions. Through analysis and simulation, we have shown that our approach has remarkable advantages in minimizing the bandwidth consumption of VOD service on CANs, compared with the previous approaches. As a result, our approach can satisfy more simultaneous VOD sessions on a fixed-capacity channel.Furthermore, we design two adaptation algorithms which not only keep bandwidth consumption minimal but also significantly reduce the mean service delay to 30% or less of the mean service delay without adaptation.Item Server-Based Dynamic Server Selection Algorithms for Enterprise Networks(2000-12-05) Dong, Yingfei; Zhang, Zhi-Li; Hou, Y. ThomasDynamic server selection is one of basic approaches to improve the quality of distributed network services over the Internet. Even though several mechanisms have been proposed, they have a number of drawbacks when applying them to an enterprise network. The authors proposed a server-based mechanism to help clients find better servers for its requests. Several server selection algorithms are proposed in this paper and tested with simulators, which are built on LBNL Network Simulator Version 2. The simulation results show that the mechanism and the algorithms are very promising. Compared with previous approaches, our approach is the first one which emphasizes both of passive measurement and server-based.Item Video Streaming across Best-Effort Wide-Area Networks with Controlled Quality Assurance Using Proxy Servers(2000-12-05) Zhang, Zhi-Li; Dong, YingfeiVideo streaming across wide-area backbone networks using proxy servers is an important component of emerging global multimedia content delivery networks. In this paper we propose and develop a novel proxy-assisted video streaming technique---referred to as the staggered two-flow video streaming technique---for delivering videos across a best-effort wide-area backbone network from a central server to a video proxy server. The objective of the proposed video streaming technique is to provide controlled video quality assurance by taking advantage of the disk space at the proxy server, while circumventing the disk I/O bandwidth limitation at the proxy server in the mean time. Our technique is designed for stored videos that use compression schemes with inter-frame dependency, such as MPEG. Using theproposed staggered two-flow video streaming technique, a video stream is delivered in two separate substreams (referred to as flows): a flow containing the essential portion of the video (e.g., I frames) and the other containing the less essential portion of the video (e.g., P/B frames.) Both flows are segmented, and transmitted across the best-effort wide-area backbone networkusing different mechanisms. By delivering the I frame flow one segment ahead of time and caching the segment at the proxy server, we use a modified TCP with application-level throughput control (called controlled TCP or cTCP) to ensure the reliable transmission of the I frame segments while meeting their delivery deadlines. In contrast, the less essential flow containing the P/B frames is streamed unreliably (using RTP/UDP) in real time, and merged with the cached I frames by the proxy server using its memory buffer. We also develop rate adaptation and control schemes to dealwith transient and persistent network congestion. Simulations are conducted to demonstrate the efficacy of the proposed staggered two-flow video streaming technique.