Industry Insights: How the Dynamic Media Facility could reshape infrastructure

As broadcasters continue exploring software-defined infrastructure, attention is increasingly shifting from individual technologies to the broader operational models that will govern future media facilities.

The dynamic media facility (DMF) concept aims to provide a framework for orchestrating media workflows across software-based environments, bringing together compute, networking, control and media services into a more flexible and scalable architecture. Yet significant questions remain around interoperability, orchestration, integration with legacy systems and the practical realities of deployment.

In the second installment of this two-part Industry Insights roundtable, participants discuss the opportunities and challenges of dynamic infrastructure, the evolving relationship between MXL and DMF, and what must happen next for these concepts to move from industry discussion to widely adopted production environments.

Key takeaways from this Industry Insights roundtable

• Architecture is shifting: Participants said DMF is encouraging organizations to think beyond fixed-function hardware and toward software-defined, dynamically orchestrated workflows.
• Orchestration matters: Much of the value in DMF comes from workflow orchestration, resource management and operational flexibility rather than any single transport or exchange technology.
• Legacy integration remains critical: Broadcasters are expected to adopt DMF incrementally, requiring software-native workflows to coexist with existing SDI and SMPTE ST 2110 infrastructure.
• Interoperability is essential: Open interfaces, shared frameworks and multi-vendor compatibility were repeatedly identified as necessary for long-term success.
• Standards need maturity: Participants said broader adoption will depend on clearer specifications, proven deployment models, workforce training and repeatable interoperability across vendors.

How is the dynamic media facility (DMF) model influencing how organizations think about system design and orchestration?

Ian Wagdin, VP, technology and innovation, Appear: DMF is encouraging broadcasters to rethink production architecture around distributed software services rather than fixed-function hardware. Critically, orchestration in this model operates at a workflow level — operators don’t just need to connect boxes but need to consider the complete workflow across multiple functions. The business case is equally concrete: by the end of 2026, major broadcasters are expected to move beyond seeing DMF as a concept and will start to utilize DMF as a measurable business engine with quantifiable cost avoidance and new revenue from microservices-driven infrastructure.

John Mailhot, SVP, product management, Imagine Communications: The DMF model is shifting the industry’s focus toward workflow deployment (orchestration), resource management, and workload flexibility across the entire media operation. The Joint Taskforce on Dynamic Media Facilities (JT-DMF) is concentrating on higher-level coordination layers above transport and media interchange technologies, including planning, timing management, and dynamic allocation of compute resources. As a result, organizations are increasingly applying cloud and data center operating principles to media infrastructure design across MXL, ST 2110, and mixed environments.

Chris Scheck, head of marketing content, Lawo: The idea of the dynamic media facility is to drastically reduce processing hardware real-estate by running the processing software on COTS servers. This allows operators to start and stop apps they do or don’t need for the upcoming project, and this build-up/tear-down strategy can be performed several times a day with a multitude of processing apps. It can even be scheduled and automated, complete with reserving the required CPU capacity and funding (credits needed to run all software-based services) using software such as DataMiner.

Paul Briscoe, chief architect, TAG Video Systems: Today’s broadcast systems are built for fixed purposes: ground, cloud, or hybrid. DMF breaks from this by constructing workflows from resources that are persistent, on-demand, or mixed, orchestrated per job and released when done. The result is granular cost management, genuine OpEx/CapEx flexibility, and visibility into production costs that are usually hidden, optimizing both resource fit and overall spend.

Adam Marshall, CPO, Grass Valley: DMF is helping shift the conversation away from individual products and towards the operating model of the facility itself. It’s encouraging organizations to think more holistically about orchestration, control, observability and security across a multi-vendor environment, rather than treating each workflow as a separate integration project. That is a healthy move, because the industry challenge now is less about whether software-defined media is possible and more about how it is operated at scale.

Advertisement

Russell Trafford-Jones, industry engagement manager, Techex: The EBU’s DMF model is moving the conversation from fixed hardware estates to modular software components that can be composed, scaled and orchestrated on demand. In that model, MXL becomes part of the connective tissue: a way of making open interchange and dynamic service creation feel practical, not just architectural. The design conversation moves from “what boxes do we own?” to “what services do we need?”

Ken Smith, senior media design engineer, Diversifed: DMF is influencing organizations to think about media infrastructure more like traditional IT and cloud environments, where workflows become software-defined, modular, and dynamically orchestrated rather than tied to fixed-function hardware. It’s accelerating the shift from FPGA-based, on-prem systems toward COTS, cloud, and hybrid processing models that can scale more elastically based on operational demand. That changes not only how systems are deployed, but also how facilities approach orchestration, resource allocation, and long-term infrastructure design.

Drew Martin, head of video product management, Riedel Communications: For many organizations, the DMF model is still fairly experimental. However, but its foundation in standard IT practices is already influencing how systems are designed and orchestrated, and it’s making collaboration with IT-focused companies more straightforward.

Francesco Scartozzi, VP, sales and business development, Matrox Video: DMF is encouraging organizations to think less in terms of fixed hardware islands and more in terms of modular, orchestrated media services. It shifts the focus toward scalable software-defined infrastructure where compute, transport, and processing resources can be dynamically allocated. This is driving broader conversations around abstraction, orchestration, and operational agility.

Where does DMF clarify system architecture, and where does it introduce new complexity or ambiguity?

John Mailhot, SVP, product management, Imagine Communications: It helps clarify architecture by defining how different layers of the design — transport, media interchange, orchestration, and compute management — interact within a modern software-defined facility. That layered framework gives organizations a clearer way to map operational responsibilities across infrastructure, applications, and control systems. At the same time, coordinating those layers introduces new complexity around interoperability, timing management, orchestration frameworks, and operational ownership.

Adam Marshall, CPO, Grass Valley: DMF, as a reference architecture, strives to give vendors and users a clearer, unified direction for how software-based platforms should be designed, ultimately improving interoperability from the outset. There is naturally still ambiguity, and the overall model is complex, but that is the purpose of the working groups: to define, validate and solve these challenges.

Olivier Suard, vice president, marketing, Nevion: At this stage, DMF is still a concept, with many details still missing, but it sets a general direction for the way software-based functionality should be developed. Any ambiguity will be resolved overtime, as DMF is defined in more detail. Far from introducing new complexity, DMF will ultimately simplify interoperability between vendor products, and therefore solution-deployment and the management of production workflows.

What are the biggest integration challenges when aligning MXL and DMF concepts with legacy systems?

Ian Wagdin, VP, technology and innovation, Appear: The biggest challenge remains bridging synchronous broadcast workflows with asynchronous, software-driven models. There is also a talent dimension; moving to software-defined workflows requires upskilling engineering teams and attracting software engineers from adjacent industries, which progressive broadcasters are already treating as a deliberate strategy. Perhaps the biggest challenge in an asynchronous environment is accurate timing and identification of media assets throughout the software-based workflow.

Chris Scheck, head of marketing content, Lawo: Using legacy equipment alongside DMF-based processing equipment is probably what most users will do at first. As long as that equipment comes with SDI or other supported connectivity, it can be connected to ST2110-compatible IP gateways, allowing acquired or output signals to be routed to the DMF as well as the vision mixer, the audio mixing console, and so on. While this does require SMPTE ST2110, it allows broadcasters to sweat their legacy (or baseband) hardware while simultaneously leveraging DMF principles and MXL for ultra low-latency media exchange on servers or in the cloud where this makes sense.

Paul Briscoe, chief architect, TAG Video Systems: It’s more than alignment, it’s tight integration that’s required. This in turn will involve bringing forward necessary service capabilities such as timing and NMOS into this new world. NMOS itself is a natural fit, but the tight synchronous system timing used in 2110 requires evolution away from 2110’s “marching band” synchronicity to a more “jazz combo” method in this new relaxed-timing world.

How are vendors supporting — or complicating — interoperability within MXL and DMF frameworks?

Jan Helgesen, head of products and solutions, Nevion: The development of MXL is managed by the Linux Foundation, with vendors contributing code to create a unified SDK. This method differs significantly from traditional standards-based approaches to defining interoperability. It is generally faster and reduces ambiguity in interpreting standards. The extent of influence over the specification correlates with the volume of code contributed. While this may favor larger organizations with greater resources, it also offers opportunities for smaller entities to innovate and have an impact.

Ken Smith, senior media design engineer, Diversifed: Vendors are supporting interoperability within MXL and DMF by consistently positioning their product lines as software applications supporting virtualized platforms and publicly support interoperability initiatives. However, vendors can have independent architectural philosophies, interoperability initiatives, and implementations can often vary beneath the surface.

Advertisement

Francesco Scartozzi, VP, sales and business development, Matrox Video: Vendors supporting MXL and DMF are focusing on open interfaces, multi-vendor interoperability, and avoiding proprietary ecosystems that recreate the limitations of traditional broadcast infrastructure. Solutions like Matrox Origin Fabric complement MXL by extending MXL-compatible media services into scalable distributed deployments across on-premises, edge locations, and cloud environments.

What needs to happen next for MXL and DMF to move from industry conversation to adopted infrastructure?

Ian Wagdin, VP, technology and innovation, Appear: Current deployments are concentrated in tier-one broadcasters and large sports productions. Mass adoption and the investment and scale that will genuinely transform the industry requires reaching tier-two and tier-three organizations. Training must also be front-loaded to ensure a smooth transition.

Chris Scheck, head of marketing content, Lawo: MXL being a necessary building block for an open, vendor-agnostic DMF approach, the first step is obviously to ensure that all apps run efficiently on generic servers, on the ground or in the cloud. To this end, Lawo’s HOME Apps have been designed as hardware-agnostic, containerized microservices that meet all DMF requirements. While their internal architecture may differ from apps developed by other vendors, all apps need to be able to run side by side and use MXL’s shared memory layer for offloading and ingesting media.

Paul Briscoe, chief architect, TAG Video Systems: MXL is the simpler piece: a media movement scheme that works standalone, with strong vendor adoption and successful interop testing. DMF is where the bigger shift happens, enabling dynamic composition of on-demand infrastructures that optimize resources and costs across both MXL and legacy methods. MXL’s real power isn’t as a transport layer in isolation, it emerges when combined with DMF’s resource orchestration to build fluid, best-of-breed solutions.

Adam Marshall, CPO, Grass Valley: In many respects, this is already happening, with production systems in the field using the same underlying concepts today, including DMC Production in Europe and CBS Sports in the US with GV AMPP. Those solutions will continue to evolve as the DMF and MXL definitions and capabilities mature. The truthful answer is that DMF needs to keep moving from discussion to definition, which is already happening at pace, and the industry is now naturally looking for more validated, repeatable implementation models around it.

Olivier Suard, vice president, marketing, Nevion: DMF and MXL have momentum behind them because the industry is realizing that COTS-based software solutions (on-prem or in the cloud) are rapidly becoming viable for live production, and deliver substantial benefits, not least in terms of flexibility. Without industry standards, there is a real danger that silo ecosystems would continue to develop, locking in customers and potentially adding risk. Vendors have embraced DMF and MXL because they realize that the industry support is behind these frameworks. What needs to happen now is that DMF and MXL need to be specified in more detail and fast.

Russell Trafford-Jones, industry engagement manager, Techex: The next step is moving from impressive demos to repeatable deployments with multiple vendors. A key technical milestone is releasing RDMA-based support, already largely developed, for extending MXL beyond a single host and scaling uncompressed software workflows across distributed compute clusters.  Alongside that, the industry needs mature orchestration, consistent monitoring and proven interoperability between independently procured vendor products, because that is what turns a good idea into infrastructure rather than another clever booth demo.