A transport layer for live streaming in a content delivery network

Abstract

Abstract-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.Streaming Content Delivery Networks 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 Content Delivery Network 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.

Quality-Optimized and Secure End-to-End Authentication for Media Delivery

“…More sophisticated acknowledgement schemes can reduce this required bandwidth to close to R S /(1-e) (depending on, e.g., constraints from the playout deadlines), however, we keep the current approach for conceptual simplicity. When channel bandwidth drops below R S /(1-e) 2 , the PSNR of authenticated video starts to drop, which is validated by all R-D curves provided. For example, in Fig.8(a), the source rate is 158kbps including 150kbps for video data and 8kbps for authentication overhead, therefore at a loss rate of 0.03 the knee of the R-D curve of RaDiO_Butterfly_Aware is located at 158/(1-0.03) 2 =168kbps.…”

“…As a further observation to understand the plots, from the sender's point of view, the channel capacity is (1-e) 2 R C , where e is the packet loss rate and R C is the channel bandwidth, because the sender considers a packet as successfully delivered only after the packet is acknowledged by the receiver. Therefore, to transmit all packets at source rate R S , the required bandwidth is R S /(1-e) 2 .…”

“…Therefore, to transmit all packets at source rate R S , the required bandwidth is R S /(1-e) 2 . More sophisticated acknowledgement schemes can reduce this required bandwidth to close to R S /(1-e) (depending on, e.g., constraints from the playout deadlines), however, we keep the current approach for conceptual simplicity.…”

“…INTRODUCTION EDIA communication over heterogeneous networks is continuing to increase in practical importance, enabled by the rapid growth of network bandwidth, improved compression formats [1], and advanced delivery technologies such as content delivery networks [2] and peer-to-peer systems [3][4] [5]. This is also evident in many commercial services and applications like IPTV (i.e., Internet Protocol Television), multimedia messaging, video conferencing, video surveillance, and so on.…”

Video-on-Demand Equipment Allocation

“…Nonetheless, the mathematics and the model are easily extended to the client-cached scenario. The CDN delivery of streaming media causes new problems that did not apply to the distribution of HTTP objects: streaming objects are much larger than web objects and hence create much more traffic [10]. Furthermore, it is no longer possible to assume infinite storage size at the replica locations [15].…”