An active router architecture for multicast video distribution

Abstract

--Video distribution over the Internet poses many challenges. Due to the best-effort nature of today's public data networks, end system applications cannot rely on either bandwidth or delay guarantees. We designed and implemented a prototype of a multicast video distribution architecture involving knowledgeable active routers, a scalable video codec based on wavelet transformation, and a high-performance video scaling algorithm implemented as a router plugin. The plugin scales the video with an average overhead of only 22 µs per video datagram and is installed on-the-fly on the routers after the sender starts transmitting video for the first time. Through experiments on our test network, we show that we can dramatically improve the video quality on the receivers (up to 15 dB PSNR) by scaling the video on the routers to almost any target bandwidth. The target bandwidth is evaluated by the router solely based on monitoring of the load situation of the router's downstream links and can be adjusted within 50 ms.

Quality of service provisioning for composable routing elements

“…A media scaling service is reported in [4] for router plugins [1]. The service applies to wavelet-encoded real-time video consisting of a base layer and progressive enhancement layers.…”

“…The service can be fetched and loaded on demand, in response to user requests. While the same service has been demonstrated in [4], our goal is to understand how resource management in CROSS/Linux can impact video quality perceived by end users. In particular, video scaling requires sufficient CPU cycles to be effective.…”

Adaptive applications over active networks: case study on layered multicast

“…In [23], a video plugin designed for multicast video distribution is built over a high performance active node platform. The video plugin is loaded on-demand using a code caching scheme, and runs in kernel space in order to achieve high performance.…”

A cluster-based active router architecture supporting video/audio stream transcoding service

“…Ralph Keller et al proposed an active router architecture [5], where video scaling algorithms are deployed to improve video performance. The difference between this model and the journey network model is that the former requires that all the scaling computation should be accomplished within the same active router, while the latter focuses on distributing computation to different routers even when the packets belong to the same stream.…”

“…In our experiments, we adopt the above function ´ Ú µ, and define the load indicator ´Øµ as ´Øµ ØØ × ¡Ø (5) where, ØØ × is the CPU time spent by the transcoding services during the polling interval ¡Ø. ´Øµ is defined as ´Øµ Ǽ…”