All 6 articles in this series are now available in our free eBook ‘Content Operations In The Multi-Screen Age’ – download it HERE.

All articles are also available individually:

Like the Set Top Box (STB), the Master Control Room (MCR) has survived almost continual predictions of its demise throughout the history of digital TV.  Also like the STB, it has had to move, sometimes reluctantly it has seemed, with the times. And those times are a changin’ right now, driven by the inexorable penetration of software defined systems and cloud compute architecture into traditional broadcasting domains.

The MCR has been a stalwart of broadcasting almost since its inception, and to date advances have mostly been internal, in terms of equipment capabilities and operating efficiencies. The changes coming now are more fundamental and could in time reshape the way broadcasting is done. There is movement towards greater outsourcing of capabilities and critical functions in terms of quality and content management, and even with sharing of resources among broadcasters.

It has been a long time in gestation, given that MCR cloudification was first mooted at least a decade ago. By 2018 such vapor had condensed into coherent steps set out by a number of traditional vendors, who started to outline plans for coordinated movement of linear and OTT master control playout to the public cloud. Companies began to develop plans to unify linear and streaming playout in the public cloud under common MAM (Media Asset Management) as a prelude for running hundreds of channels that way 24/7, including live events, time delayed, and rebroadcast content.

Overcoming The Challenges

In the event, this was jumping the gun, and even as recently as 2023 seasoned MCR engineers were warning that the public cloud was not quite ready to handle the complex, dynamic and real time requirements of local stations. Some of the criticisms were common to other industry sectors, including being dependent on internet access and public cloud providers, as well as being at the mercy of whatever bandwidth was available and being susceptible to connectivity failures. This is especially applicable at times of natural disasters such as an earthquake which is precisely when people rely on broadcast services. 

To some extent these common deficits of the public cloud have been overcome. There was a time when it was thought too risky to rely at all on internet access for critical applications under the banner of “network computing.” Later there was kickback against cellular networks for emergency calling.

But as mobile services and broadband connectivity generally matured, many of the old deficits faded away. That maturation process has now extended to fields just as demanding in terms of bandwidth, robustness and availability as broadcasting. No longer is streaming considered a second-hand offshoot of linear delivery, destined to remain a poor relation in terms of quality. Instead it now represents the future of prime-time delivery for a growing number of broadcasters. 

Migration To Software Defined Infrastructure

Inevitably then, the time would come for MCR to migrate to more flexible and scalable software defined infrastructure. In that case there has also been a constraint specific to broadcasting in the old school practice of central casting, sometimes known as hubbing. This is where MCR equipment and the people running it are concentrated in centralized or regionalized technical centers supporting multiple stations.

This has left a number of TV stations continuing to invest in on-premise hardware for master control, while initially focusing cloud compute deployments on streaming, playout of recorded programming, and ads. The public cloud has also featured increasingly for disaster recovery, achieving a greater degree of geographical redundancy. This minimizes risks associated with natural disasters such as earthquakes, while also serving as a pivot towards cloud compute resources.

As a result, most of the major providers of broadcast technology have continued to offer both on premises and public cloud options, catering for on-premises, private, and public cloud environments, or hybrids of these.

The tide towards cloudification continues to flow as the supporting logic gathers force. This is witnessed in the composition of the IBC Accelerator Projects for its 2025 Accelerator Media Innovation Program culminating in this year’s event in September. Of the eight Accelerator projects and one Special Incubator project selected as finalists, one is dedicated to master control in the public cloud, while several others have some relevance for it.

The dedicated accelerator is called Master Control Cloud and was proposed by the BBC, Sweden’s national pubcaster SVT, and BT Media & Broadcast. The focus is on incorporating traditional MCR functionality in live media streams that would then be processed in the public cloud.

Talkback

A significant aspect of this project was the observation by all three of these broadcasters that there was a strengthening pull towards cloudification from within some of the constituent functional groups. This was particularly evident in broadcast talkback systems, which are part of the intercom systems providing secure and low latency communication between control and recording rooms. They comprise microphones, receivers, headsets, as well as software controlling switching and audio quality.

Intercom and constituent talkback systems have themselves been evolving towards IP audio transmission and adoption of wireless communications, especially 5G. They have also incorporated AI just as the way software has done for applications that involve adaptation through data analysis. This is tilting intercom applications towards virtualized cloud, which allows them to be based on more cost effective and flexible COTS (Common Off The Shelf) hardware and be managed remotely. Cloud compute based intercom and talk back systems are therefore gaining traction themselves as part of the bigger MCR picture, leading to pressure from within to move that way.

At this level too there is a market for hybrids with the emergence of virtual intercom and talkback platforms that can be deployed in the cloud, or on servers on premise, or combinations of the two. The decision over which to deploy at this stage will depend on where the production teams are. With increased remote production that is pivoting towards the cloud, but where production teams are still in the control room, on premises systems might be preferred, not least because that minimizes signal latency. It comes down to workflow in the end.

It may not yet quite be a stampede, but the cloudification march is gathering pace and force from its own momentum. As broadcasters that have already embarked on migration obtain competitive advantage where that is relevant, the urgency for others to follow increases.

Scaling Up

The advantage is clearest for bigger broadcasters with many channels. These benefit most from much more dynamic scalability of the signal chain when MCR has been cloudified. Now resources can be turned on and activated within software defined infrastructure almost automatically in the event of a new channel coming on stream, or equally can be decommissioned at a stroke when a channel is shut down.

This is possible because the whole chain of signaling, from ingest through transcoding to final playout, along with associated steps such as real time content moderation, can scale dynamically. They no longer have to be installed as physical systems but instead orchestrated in software.

There is a caveat and that is in the composition of the cloud compute platform. The public cloud is sometimes erroneously considered to be a bottomless pit of infinite scalability and elasticity. In reality it is a finite set of interlinked virtualized servers and storage systems distributed across multiple locations, which can be privately owned by a single enterprise or publicly provided by say a hyperscaler like AWS, Microsoft Azure or Google Cloud.

The actual degree of scalability, elasticity and redundancy depends in part on the extent of the infrastructure, which is why so many enterprises default to one of those big hyperscalers. That is happening in broadcasting because broadcasters know the capacity will be there when they need it.

Collaboration Is Key

But video services have some specific requirements, and that is leading some broadcasters to collaborate over their own infrastructures, even if these are underpinned by the major public cloud platforms. The most notable example so far is at the European Broadcasting Union (EBU), given that this has 113 members in 56 countries with an additional 31 Associates in Asia, Africa, Australasia and the Americas. It therefore has the scale for big inter-broadcaster projects.

The EBU is currently engaged in a major digital transformation project called MARS (Multilayer, Anywhere, Resilient and Sustainable) based on its Dynamic Media Facility (DMF). The latter is a scalable infrastructure allowing transparent switching between local servers, private data centers, sovereign national clouds and public clouds. 

The EBU’s long term vision is to develop a European media cloud shared among all members, which would enable broadcast centric applications, including MCR functions, to run on common infrastructure for better scaling at that level. These would run back to public clouds, probably featuring all three major hyperscalers.

A core enabler will be the MXL (Media Exchange Layer), defining how assets are exchanged between media functions. This is work in progress, and also featured in one of 2025’s IBC Accelerator Projects.

That is the Multi-Vendor Software Live Media Exchange, proposed by the EBU with the BBC and Norwegian media and a growing community of content technology companies. This revolves around the DMF, with the aim of leading broadcasters towards modern distributed data centers and cloud-based infrastructures.

In the rest of this mini-series we will probe more deeply into the shifting sands of content operations, taking cues from some of the leading broadcasters and video service providers in the field, while discussing challenges and constraints in greater detail.

All 6 articles in this series are now available in our free eBook ‘Content Operations In The Multi-Screen Age’ – download it HERE.