[Paper Review] Transcoder Migration For Real Time Video Streaming Systems

Paul Farrow, Martin J. Reed+2 more|arXiv (Cornell University)|Sep 27, 2015

Video Coding and Compression Technologies9 references3 citations

TL;DR

This paper proposes a dynamic transcoder migration system for real-time ultra-high definition (4K) video streaming using OpenFlow-enabled networks. By combining a heuristic algorithm for optimal transcoder placement with OpenFlow-assisted migration, the system enables near-seamless transcoder relocation during active streaming, reducing client-side disruption to sub-second delays and significantly improving QoS under dynamic network conditions.

ABSTRACT

The increase in real time ultra-high definition video services presents a challenging issue to current network infrastructures, because of its high bandwidth usage, which saturate network links. The required bandwidth is related to strict QoS requirements for digital media. There are systems in place currently to help reduce these problems, such as transcoders and application layer multicasting. However, these approaches are limited because they are usually implemented as static resources. In contrast, by using the OpenFlow based system presented in this paper, it is possible to provide a more effective approach using dynamic resources - by both optimally placing transcoders in the network, as well as by migrating them to different locations while the streaming is taking place. This migration mechanism provides a near seamless switchover with minimal interruption to the clients.

Motivation & Objective

• To address the challenge of high bandwidth consumption in real-time 4K video streaming over constrained network links.
• To overcome the limitations of static transcoder and multicast systems by enabling dynamic, on-demand placement and migration of transcoders.
• To minimize client-side disruption during transcoder migration through OpenFlow-based traffic redirection.
• To optimize network resource utilization by placing transcoders adaptively based on client distribution and load.
• To evaluate the performance of OpenFlow-aided migration against standard migration methods in inter-continental 4K streaming scenarios.

Proposed method

• A heuristic algorithm is used to optimize transcoder placement, minimizing total network load while maintaining QoS.
• OpenFlow-enabled switches are used to dynamically redirect traffic during transcoder migration via flow table modifications.
• The system leverages virtual machines in cloud datacenters to enable flexible, scalable transcoding resources.
• Traffic migration is managed through a control software layer that issues OpenFlow FlowMod messages to update switch forwarding rules.
• Application layer multicasting is integrated with transcoders to reduce redundant traffic and improve bandwidth efficiency.
• The migration process is tested using inter-continental 4K video streaming with real-time performance monitoring of packet delays and client disruption.

Experimental results

Research questions

• RQ1 How can transcoder placement be optimized dynamically to reduce network load in response to shifting client distributions?
• RQ2 What is the impact of OpenFlow-assisted migration on client-side video disruption during transcoder relocation?
• RQ3 How does the performance of OpenFlow-aided migration compare to standard migration in terms of delay and stream continuity?
• RQ4 To what extent can dynamic transcoder migration improve QoS and bandwidth efficiency in real-time 4K streaming?
• RQ5 Can the system maintain high performance while reconfiguring transcoder locations during ongoing streaming sessions?

Key findings

• OpenFlow-aided transcoder migration reduced client-side disruption to sub-second delays, with minimal visible impact on video playback.
• Standard migration methods caused significant delays and video artifacts, including freezing and frame loss, due to prolonged stream re-establishment.
• The heuristic algorithm achieved transcoder placement results comparable to a genetic algorithm but in a fraction of the time, enabling frequent re-optimization during streaming.
• The use of transcoders as application layer multicast points significantly reduced overall network traffic, especially over long-haul links.
• The system demonstrated near-seamless migration with 95% confidence intervals showing negligible delay spikes during OpenFlow-assisted transitions.
• The control software successfully managed flow table updates across OpenFlow switches, enabling precise and low-latency traffic redirection during migration.