Media Functions Virtualization: the reference architecture

Posted on: August 14, 2019 by Mauro Starinieri

In previous blogs I’ve talked about the important technology transformation that is happening in the broadcast media industry: the move away from proprietary and bespoke broadcasting hardware technology to software running on commodity IT hardware, that could also be virtualized, and the transition from Serial Digital Interface (SDI) baseband to IP video routing and networking. This shift is inevitable and is already changing the broadcasting infrastructure across the content supply chain.

The main advantages of adopting IP and virtualized IT infrastructure in broadcasting are flexibility, agility and reduction of TCO. Implementing a change in the media supply chain for introducing: a new channel, a new content format or setting up the resources for the next live show, now consist of reconfiguring or starting the appropriate software applications, rather than installing and/or moving hardware boxes.

This sounds simple but like all things, appearances can be deceiving. To start with, SDI deals with uncompressed content which requires large bandwidth but ensures a high-quality real-time content, something that IP struggled to replicate until recently. Moreover, IP networking and enterprise compute provide large opportunities to broadcasters, but with this comes a set of new challenges such as: cybersecurity, IT provisioning, system monitoring, virtualization and ultimately controlling the effective use of cloud compute.

The Joint Taskforce on Networked Media (or JT-NM) since 2015 has defined a multi-layer reference architecture for the transition from purpose-built broadcast infrastructure to standard IT/IP infrastructure.

In a similar way, we can slice the virtualized and IP-based broadcasting infrastructure into four different layers:

1. The computing, storage and networking hardware infrastructure. This is commodity IT hardware based on different computing platforms (CPUs, GPUs, FPGAs), that can be deployed on-prem in your own data center or externalized in a private or public cloud.

2. The virtualization layer: it consists of the software stacks that implement the virtualization of the underlying hardware infrastructure of the first layer. Also, this virtualization layer can be implemented in your own data center, or in a private or public cloud.

3. These first two layers realize the virtualized broadcast data center, which provides the virtualized resources (computing, storage and networking) upon which the software of the virtualized media functions is deployed and executed to implement the related media workflows of the content supply chain. Those virtualized media functions are provided by specialized vendors or are accessible from the media services catalogs that public cloud service providers like AWS, Azure, and Google make available as-a-service to their customers in addition to the basic cloud infrastructure.

4. On top of everything, the analytics, mainly real-time. We used to think about analytics in media mostly in the context of audience measuring and profiling and content recommendation. This is still very important for media companies, but with the broadcast media processes increasingly moving to IP/IT infrastructure, much more analytics, on real-time basis, can be applied by collecting huge amounts of data about infrastructure, networks, applications, processes and services. Many interesting additional use cases are enabled, such as: predictive maintenance, process optimization, quality of experience and quality of service real-time monitoring, etc.

Cybersecurity is a key transversal area across these four layers. It is a very critical topic and a real priority for all the industries and particularly for the broadcasting and media industry. With the growing adoption of IT and IP, the broadcasting industry is today much more exposed to cybersecurity risks. It’s not only a matter of content protection which is of course still critical for the media business, but it’s also a matter of business and service continuity as well as customers’ data protection. The good news is that in cybersecurity we can apply the full portfolio of services, solutions and best practices that are already deployed in the Enterprise IT and Operation Technology contexts of other industries.

The broadcasting infrastructure based on IP/IT has, indeed, many advantages, as specified above. However, it is also quite complex and can work efficiently only with the proper automation and orchestration of the provisioning and configuration of the required physical and virtualized resources, and of the applicative functions across the media workflows. This is why the management and orchestration are the intelligent core of any new IP/IT-based broadcast infrastructure.

Finally, we cannot forget the legacy broadcasting technology, because it will take years to replace it. We need to integrate it and to guarantee its interoperability with the new IT/IP-based infrastructure, from a technology and an operational point of view.

It is then clear that a new mix of capabilities and skills is required for planning, designing and executing a successful transformation roadmap to IP/IT and virtualization, that can maximize the benefits of the new technologies, without compromising the level of performance, continuity, quality and security required by the media workflows and processes.

Industry associations like AIMS have started promoting and facilitating broadcast’s transition to IT/IP. The IPShowcase at the NAB and IBC tradeshows is one of its initiatives, involving also other associations (SMPTE, AMWA, AES, VSF, EBU) and many vendors, to educate attendees on the business and creative potential of IP-based media solutions.

AMWA started implementing proof-of-concepts, best practices, guidelines and reference architectures and interoperability specifications with the support of key vendors and service providers.

IABM and DPP are also sponsor of initiatives around the transition of broadcast chain to IP/IT and virtualization.

For more insights, we’ll be showcasing practical expertise on how we help broadcast and media companies meet their enterprise IT challenges. Meet us on our booth from 13-17 September at IBC 2019. More info

Share this blog article

  • Share on Linked In

About Mauro Starinieri
Head of Global OSS/MANO Competence Center
Mauro has more than 30 years of experience in the IT, telecommunication and media industry. With more than 11 years of experience in telecom OSS domain, he is currently leading the global OSS and Management & Orchestration (MANO) competence center, supporting a wide variety of telcos in transforming their operation support applications for the digital services and the next generation networks. He is responsible for building, managing and promoting the Atos OSS/MANO solutions to drive the growth of Atos business around the OSS/MANO domain. Mauro has also served as global portfolio manager for the media market. In his career he played many different roles in technical, business development and sales areas. Before joining Atos, he worked at Siemens, HP and IBM.

Follow or contact Mauro