Algorithms for advance bandwidth reservation in media production networks

Media production generally requires many geographically distributed actors (e.g., production houses, broadcasters, advertisers) to exchange huge amounts of raw video and audio data. Traditional distribution techniques, such as dedicated point-to-point optical links, are highly inefficient in terms of installation time and cost. To improve efficiency, shared media production networks that connect all involved actors over a large geographical area, are currently being deployed. The traffic in such networks is often predictable, as the timing and bandwidth requirements of data transfers are generally known hours or even days in advance. As such, the use of advance bandwidth reservation (AR) can greatly increase resource utilization and cost efficiency. In this paper, we propose an Integer Linear Programming formulation of the bandwidth scheduling problem, which takes into account the specific characteristics of media production networks, is presented. Two novel optimization algorithms based on this model are thoroughly evaluated and compared by means of in-depth simulation results.