Provisioning content distribution networks for streaming media

2009 International Conference on Computer Technology and Development, 2009

To deliver media streaming applications efficiently, researchers have developed three practical mechanisms, each of which has its virtues and limitations: Infrastructure-based CDN, Proxy-based approaches and P2P Systems. To address some of the above system limitations we present an enhanced media streaming design. Our design organizes its network entities in a structured P2P manner to provide a large media storage and to dynamically participate in the media streaming delivery, where overlapping clusters, Proxy based, are to minimize delay of streaming service. Multiple servers besides offering increased resources and service availability allow a further reduction of network bandwidth usage. The challenge is how to use multiple servers efficiently to address both the scalability issue in centralized proxy systems and the unreliable service concern by only depending on the clients peering & proposing a distributed architecture that can timely serve the demanded media data and fairly allocate media streaming tasks across the system. Finally, an algorithm is introduced showing that the proposed scheme significantly reduces the workload posed on the origin server.

IEEE International Conference on Networks, 2003

The current paradigm of content distribution networks (CDN) deployment requires heavy infrastructure investment since a large number of servers have to be deployed over a wide area. To overcome this difficulty, we propose a new paradigm where future CDNs are to be deployed over a leased server infrastructure. This paradigm shift changes the object placement problem in current CDNs to

Proceedings of The IEEE, 2004

Streaming media on the Internet has experienced rapid growth over the last few years and will continue to increase in importance as broadband technologies and authoring tools continue to improve. As the Internet becomes an increasingly popular alternative to traditional communications media, Internet streaming will become a significant component of many content providers' communications strategies. Internet streaming, however, poses significant challenges for content providers, since it has significant distribution problems. Scalability, quality, reliability, and cost are all issues that have to be addressed in a successful streaming media offering. Streaming content delivery networks (streaming CDNs) attempt to provide solutions to the bottlenecks encountered by streaming applications on the Internet. However, only a small number of them has been deployed, and little is known about the internal organization of these systems. In this paper, we discuss the design choices made during the evolution of Akamai's CDN for streaming media. In particular, we look at the design choices made to ensure the network's scalability, quality of delivered content, and reliability while keeping costs low. Performance studies conducted on the evolving system indicate that our design scores highly on all of the above categories.

Streaming media on the internet has experienced rapid growth over the last few years and will continue to increase in importance as broadband technologies and authoring tools continue to improve. As the internet becomes an increasingly popular alternative to traditional communications media, internet streaming will become a significant component of many content providers' communications strategy. Internet streaming, however, poses significant challenges for content providers since it has significant distribution problems. Scalability, quality, reliability, and cost are all issues that have to be addressed in a successful streaming media offering.

In this paper we present some contemporary examples of research-oriented as well as commercial content delivery networks (CDN)s. We discuss how both types of CDNs complement each other in tackling the posed technical and commercial challenges. We also show which performance metrics such as quality of experience (QoE) and time to first byte (TTFB) best capture the dynamics of traffic and services in CDNs. The core of this paper is our proposed network architecture for CDN providers/operators. The architecture combines video multicast optimized trees and cross-layer coordination between the physical DWDM layer (L1) and network layer (L3) for achieving higher efficiency and lower latency values for live streaming and on demand (VoD) video. Due to the pilot implementation of the presented concept being limited in scale, we use simulations in order to perform proof-of-concept on a sufficiently large environment. Results show that there is a strong correlation between the TTFB and QoE metrics with the former taking on values as low as 75 msec in a national 3-tier network.

IEEE Internet Computing, 2001

WCC 2000 - ICCT 2000. 2000 International Conference on Communication Technology Proceedings (Cat. No.00EX420), 2000

Communication technologies have made multimedia on demand services technically and economically feasible. An important aspect of such systems is the resource sharing technique, which allow the contemporaneous service of a large number of user requests with a considerable saving in terms of network bandwidth and server resources. In this paper, we report the results of a study which analyzes batching and buffering techniques to group and serve together video requests. The mathematical model allows the evaluation of the main system performance (probability distribution of the number of streams, percentage reduction of resources, etc.) as a function of load and batching interval duration. Simulation experiments confirm the analytical model in the whole range of considered conditions. Abstract Recent advances in Information and Communication technologies have made multimedia on demand services technically and economically feasible. An important aspect of such systems is the resource sharing technique, which allow the simultaneous service of a large number of user requests with a considerable saving in terms of network bandwidth and server resources. In this paper, we report the results of a study which analyzes batching and buffering techniques, that is grouping and serving together video requests. The mathematical model allows the evaluation of the main system performance (probability distribution of the number of streams, percentage reduction of resources, etc.) as a function of load and batching interval duration. Simulation experiments confirm the analytical model in the whole range of considered conditions.

2009

Peer-to-peer (P2P) systems are gaining increasing attention in research as well as industrial communities. In this paper, we propose a new model for on-demand media streaming centered around the P2P paradigm. The proposed P2P model can support a large number of clients with a modest overall system cost. The P2P model allows for peers to share some of their resources with the system and in return, they get some incentives or rewards. We propose two architectures to realize (or deploy) the proposed model. One architecture relies on a special entity, an index, to facilitate locating the peers in the system. The second architecture is a pure P2P and builds an overlay layer over the participating peers. For each architecture, we present new dispersion algorithms (for disseminating the media files into the system) and searching algorithms (for locating peers with the required objects).

2003

Content Delivery Networks (CDN) are designed to efficiently deliver media content from content providers to a large community of geographically distributed users. The media type varies from the Web content to audio and video streaming. CDN providers are interested in placing CDN servers at suitable locations on the Internet to serve their users with the minimum cost. In this paper, we formulated the CDN server placement problem on the public Internet as a mixed integer programming model. The numerical results show that the model can be solved within tens of seconds in an AS-level topology of about 1000 ASs using a greedy decomposition method. In addition, simulation-based performance evaluation for various placement models in simulated Internet AS-level topologies shows that our model outperforms the random model and the K-median model. Sensitivity analysis is conducted empirically to study the impact of the deviation of user bandwidth demand and the link available bandwidth upon the system performance.