Implementing Software-Defined Networking (SDN)-based traffic steering model for video on demand (VoD) services


VoD/CDN delivery ecosystem

Video on demand (VoD) continues to grow at an unprecedented rate, as millions of users subscribe to digital on-demand video viewing experience globally. Cisco predicts that the global IP video traffic will be 82% of all consumer Internet traffic by 2021, compared to 73% in 2016. Also, the consumer VoD traffic will nearly double by 2021. However, digital service providers (DSPs) face a huge challenge in delivering high-quality viewing experience to the customers due to this multi-fold increase in consumption of on- demand video content and the increased complexities associated with it. The content consumption pattern of the users is changing rapidly. They have huge expectations in terms of seamless delivery of on-demand content across devices, with a special inclination towards mobile devices. DSPs rely heavily on their own CDN cache regions or interconnects with third-party ISPs. Congestion on these links can directly affect the customer experience. The challenges posed by such complex VoD/CDN delivery ecosystem in ensuring congestion-free traffic delivery are:

  • Unpredictable service outages – link failures, CDN region failures, hardware failures
  • Inefficient utilization of capacity and resources
  • VoD traffic congestion due to inefficient routing methodologies (e.g. manual CLI-based routing configurations)

DSPs need a real-time sophisticated traffic management mechanism to deliver quality video viewing experience. SDN-based traffic steering with central management plane is the ideal solution to handle these challenges.

This insight covers the key steps involved in the implementation of the SDN-based traffic steering model for VoD traffic handling:

  • Implementing SDN controller to enable centralized control plane with a vendor-neutral network abstraction
  • Developing a traffic engineering application to manage both legacy MPLS network elements and SDN-ready elements
  • Configuring automated and dynamic routing policies on traffic engineering application to achieve sophisticated traffic steering



Praveen Chalamalasetti
Rakesh Roy
Vishwa Nigam

footer logo

Log in with your credentials

Forgot your details?