Industry Insights: Balancing flexibility and reliability in cloud workflows

As broadcasters expand cloud and hybrid production environments, operational realities are beginning to shape deployment decisions as much as flexibility and scale.

In the second installment of this Industry Insights roundtable on cloud production, vendors and technology leaders examine the engineering, reliability and workflow challenges that emerge when live production moves beyond dedicated hardware and into distributed infrastructure.

The discussion explores how organizations are balancing cloud agility with the deterministic performance requirements of live production, including latency management, redundancy, monitoring and orchestration across mixed environments. Participants also examine how containerization, microservices and IP-based architectures are changing system design while legacy infrastructure, cost predictability and operational complexity continue to influence adoption strategies.

Across the conversation, a common theme emerges: cloud production maturity depends less on infrastructure location and more on operational consistency, resilience and control.

Key takeaways from this Industry Insights roundtable

• Predictability matters: Broadcasters are prioritizing operational consistency and deterministic performance alongside cloud flexibility and scalability.
• Hybrid persists: Organizations continue to keep latency-sensitive functions on-premises while using cloud infrastructure for elasticity, orchestration and overflow capacity.
• Monitoring evolves: Resilience increasingly depends on continuous monitoring, failover logic and automated recovery rather than traditional hardware redundancy alone.
• Complexity grows: Integrating cloud workflows with legacy SDI, IP and streaming infrastructure introduces synchronization, orchestration and operational management challenges.
• Software reshapes infrastructure: Containerization and microservices are allowing workflows to become more modular, portable and scalable while reducing dependency on dedicated hardware.

How are organizations balancing flexibility and scalability against the operational predictability of traditional facilities?

Rene van Koll, senior solutions architect, Big Blue Marble: Cloud workflows enable rapid scaling to meet fluctuating demand, particularly during live events, but require more structured design to maintain consistent performance. As a result, there is greater emphasis on monitoring, latency management, and recovery strategies to ensure systems behave predictably under pressure. Live sport remains a key reference point, where remote production and multi-platform delivery must operate reliably at scale. This has led to an approach where the cloud enables flexibility, while predictability comes from established broadcast engineering practices.

Yaya Selva, CMO, Net Insight: For a few years, the industry talked about flexibility almost as if that alone was the goal. What we see from customers now is a much more mature view: Flexibility matters, but not if the behavior of the live service becomes harder to predict when the pressure is on. The real requirement is to combine cloud-scale agility with the same operational confidence people have historically expected from fixed facilities.

Ian Wagdin, VP, technology and innovation, Appear: In simple terms, organizations are keeping the parts of the workflow that benefit from deterministic performance in dedicated infrastructure, while using software and cloud where agility, orchestration and elastic scale add the most value. They can modernize without having to rebuild everything at once. The priority now is striking that balance, moving beyond cloud versus facility to a model where live production runs as a single, coherent workflow across both.

Rick Young, SVP, head of global products, LTN: Broadcasters are shifting away from fixed distribution structures toward models where a single live event can support multiple versions, formats, and endpoints without adding operational complexity. Broadcast-grade IP contribution and distribution enables this by combining dependable performance with more adaptable production and delivery workflows. Capabilities such as regionalized channel variants, tailored feeds, and support for formats like 1080p60 HDR are now being deployed in live workflows, reflecting a shift toward greater differentiation and localization.

Advertisement

Nina Walsh, global leader, business development, GTM & solutions, media & entertainment, games and sports, AWS: In many ways, the cloud provides greater operational predictability than on-premises resources, which require significant CapEx investment based on forecasted needs as well as ongoing management. When organizations understand cloud-based infrastructure and how it works, it fundamentally changes how they operate. Instead of being bound by capacity, they can think about how to meet opportunity and have access to the tools to implement those plans.

Chris Scheck, head of marketing content, Lawo: When thinking about the agility public- or private-cloud workflows can unleash, users need to understand that IP is a prerequisite to get the signals to where they are processed. This boils down to video, audio, or combined gateways to which existing SDI infrastructure can be connected. Once they are in place, the need for dedicated hardware dwindles as processing stacks on servers or in the cloud can be reshaped indefinitely. 

How are teams managing latency, reliability and signal quality when producing live events through cloud infrastructure?

Yaya Selva, CMO, Net Insight: The strongest teams have stopped assuming the cloud itself solves live performance. They treat latency, timing, path behavior and recovery as end-to-end service questions, not isolated infrastructure questions. In live, what matters is not whether the system works in principle, but whether it behaves consistently across feeds, sites and failure scenarios.

Ian Wagdin, VP, technology and innovation, Appear: It does not matter how flexible a workflow looks on paper if it cannot deliver consistent timing, stable performance and broadcast-grade quality under pressure. That is why broadcasters are being much more selective about where cloud processing sits in the workflow. They are designing architectures that support live requirements without creating unnecessary movement of signals between environments. In many cases, the most sensible principle is that once content enters the cloud, it should stay there for as long as possible.

Rick Young, SVP, head of global products, LTN: To support this, broadcasters are turning to managed IP distribution that includes continuous monitoring, built-in redundancy, and service-level assurances. These capabilities provide greater control over performance and help maintain signal integrity throughout the production chain. Broadcasters now require infrastructure that can sustain performance across variable network conditions while supporting the required flexibility of live operations, ensuring that cloud-based production environments operate with the same level of confidence as traditional models.

John Mailhot, SVP, product management, Imagine Communications: Latency is still one of the biggest considerations, especially for live production where timing is everything. Most organizations are taking a hybrid approach, keeping latency-sensitive parts of the workflow closer to the operators, while using cloud in other parts of the pipeline. Advances in low-latency codecs are helping, but in the end, it still comes down to how the workflow is designed and how operators will interact with it.

What role do containerization, microservices or other software architectures play in modern cloud production systems?

Chris Scheck, head of marketing content, Lawo: Containerization is crucial as it abstracts the processing apps from the hardware they run on, allowing for no-compromise portability between servers. Microservices, for their part, offer the advantage that they can be used in a variety of circumstances, e.g. to provide inputs and outputs in different transport formats, specific processing aspects that play a part in several apps, and so on. Thanks to containerization and microservices, apps are also much easier to maintain and upgrade, meaning that bugfixes and feature extensions can be delivered on a monthly rather than yearly basis.

Ryan Durland, cloud engineering, Diversified: By deploying discrete functional components as independent services, systems integrators can precisely tailor software workloads to the underlying infrastructure. This approach ensures organizations license only the capabilities they actively use, reducing unnecessary software spend and lowering total cost of ownership. In parallel, isolating services improves system reliability, as individual components can be restarted, scaled, or rebuilt independently without disrupting the broader environment, supporting higher availability and faster incident recovery.

How are organizations ensuring redundancy and disaster recovery in cloud-enabled production workflows?

Golan Simani, director of cloud and technical operations, TAG Video Systems: Redundancy in cloud workflows requires rethinking what backup means. In hardware environments, redundancy was physical: a second device ready to take over. In cloud and hybrid environments, it’s increasingly about failover logic, state management, and real-time health monitoring that can trigger automated responses. The monitoring layer is part of the resilience architecture rather than just an observer of it.

Rene van Koll, senior solutions architect, Big Blue Marble: Multi-CDN strategies and geo-redundant deployments are used to maintain continuity, allowing traffic to be rerouted and services to remain available during regional issues or provider degradation. A key advantage of cloud is how this redundancy is provisioned. In on-prem environments, achieving 1+1 redundancy typically requires duplicating infrastructure regardless of how often it is used, whereas in the cloud, equivalent capacity can be kept available and only activated when needed. 

Nina Walsh, global leader, business development, GTM & solutions, media & entertainment, games and sports, AWS: Cloud-enabled workflows are optimal for redundancy and disaster recovery, it’s just a matter of organizations having the right strategy in place. Audiences are not going to wait for content to be restored if infrastructure goes down so these plans are crucial. Redundancy is built into AWS infrastructure; we have at least three availability zones in each AWS Region so there are multiple backups if something goes awry.

What challenges remain when integrating cloud tools with legacy broadcast infrastructure?

Yaya Selva, CMO, Net Insight: The hardest part is usually not connecting the tools. It is maintaining timing, service assurance and operational consistency when SDI, IP, cloud and partner-operated environments all have to work together in the same live chain. That is why so many customers prefer stepwise modernization over wholesale replacement.

Advertisement

Rick Young, SVP, head of global products, LTN: Relying heavily on public cloud workflows can introduce cost unpredictability and performance concerns, particularly in complex, multi-partner environments where data transfer and service layers quickly add up. This is further complicated by the need to support both legacy distribution and rapidly growing streaming platforms. As a result, many are re-evaluating their cloud strategies and moving toward hybrid approaches that combine public cloud with private infrastructure and data center resources.

Guillaume Aubuchon, VP of product management, Avid: Legacy infrastructure wasn’t designed for dynamic, software‑driven workflows. Latency sensitivity, fixed pipelines, and siloed systems can all create friction when integrating cloud tools. The industry is responding by introducing abstraction layers and shared services that bridge old and new, allowing organizations to evolve workflows without destabilizing mission‑critical operations.

John Mailhot, SVP, product management, Imagine Communications: The main challenge is complexity — once you’re operating across multiple environments, everything has to stay synchronized. That includes content, metadata, and signal paths, all of which require strong orchestration to manage effectively. Tools that enable operators to understand the dynamic system and quickly isolate and respond to issues are increasingly important.

What operational tradeoffs are emerging between performance, flexibility and cost as cloud production scales?

Rick Young, SVP, head of global products, LTN: Traditional systems provided stability through fixed delivery paths, while cloud-based and IP-driven approaches introduced greater flexibility in how content is produced and distributed. Cloud-enabled workflows allow multiple versions of content to be delivered across many platforms from a single core feed, making it easier to support additional regional variants while reducing the need for upfront capital investment. Broadcasters are asking for fully-managed, reliable services that deliver guarantees on redundancy and availability.

Chris Scheck, head of marketing content, Lawo: What matters most is a solid strategy based on in-depth testing to spot the benefits and pitfalls before rolling out a cloud-based workflow. A clear idea about the cost of cloud ingress and egress should guide the decision-making process. Processing apps are more powerful and agile than bespoke hardware solutions, and the compute on which they run keeps evolving. As long as latency can be kept in check and cost remains predictable, nothing should stop operators from migrating to a private or public cloud.