The Media eXchange Layer (MXL) open-source framework has won the IBC 2026 Innovation Award for Content Creation. Developed by the EBU, CBC/Radio-Canada, and Grass Valley, the project provides a shared-memory architecture to enable low-latency interoperability between software-based media functions in live production clusters.
The recognition of MXL signals a shift in live production from purpose-built hardware to general-purpose compute clusters. By utilizing shared memory rather than traditional IP transport like ST 2110 for internal processing, the framework removes the latency bottlenecks that previously hindered software-defined workflows. This collaborative approach between competitors like Grass Valley and Imagine Communications suggests the industry is prioritizing a unified foundation for cloud-native and AI-driven applications. As broadcasters move away from proprietary stacks, this open architecture could lower the barrier for entry for new software vendors. Watch for the first large-scale deployments of MXL-compliant clusters in Tier 1 live sports environments to validate these efficiency gains.
The Media eXchange Layer has rapidly expanded its contributor base since its initial release, with the Linux Foundation hosting the project under its open-source governance model. In June 2026, the IEEE ComSoc Technology Blog documented a cluster of announcements from Ericsson, Nokia, and Verizon signaling a shift from AI research to commercial AI-driven network automation, a trend that parallels the broader movement toward software-defined infrastructure in media production. The same principle of replacing proprietary hardware with interoperable software layers running on general-purpose compute is driving MXL adoption among EBU members and North American broadcasters. CBC/Radio-Canada, one of the three founding contributors, has been integrating MXL into its live production pipelines as part of its broader transition away from baseband-centric workflows.
On the business and partnership side, Grass Valley has positioned MXL as a complement to its existing media processing portfolio rather than a replacement for ST 2110 transport. The framework targets the intra-cluster communication layer, where software functions running on the same compute node can exchange frames through shared memory instead of serializing and deserializing over IP. This distinction matters for broadcasters evaluating cloud-native production architectures, because it reduces the processing overhead that has historically made software-based live switching less responsive than dedicated hardware. Nokia's work with AWS and Databricks to build a unified telco data platform for autonomous networks illustrates the same architectural pattern being applied in telecom: a shared data and control layer that sits between domain-specific applications and underlying infrastructure, enabling cross-domain automation without rewriting code for each environment.
Technical validation of MXL's performance claims has come through live production trials at EBU member organizations. The shared-memory approach eliminates the serialization latency inherent in ST 2110 encapsulation for functions running within the same processing cluster, achieving frame-accurate handoffs at sub-millisecond intervals. Light Reading reported that Nokia is deploying agentic AI into its mobile core with smaller models collocated at the network edge, reducing call setup times from roughly 10 seconds to one or two seconds, demonstrating how colocated processing and reduced serialization overhead yield dramatic latency improvements in adjacent infrastructure domains. For live production, MXL's architecture enables similar gains by keeping media functions within a single memory space, a design choice that positions the framework as a foundational layer for AI-assisted production tools that require real-time access to uncompressed video frames.
The MXL open-source framework won the IBC 2026 Innovation Award for Content Creation. By utilizing shared-memory architecture instead of traditional IP streaming, MXL enables low-latency interoperability between software-based media functions. This shift allows broadcasters to move away from proprietary hardware toward efficient, general-purpose compute clusters for live production.
MXL is an open-source framework that uses local and remote shared memory to enable low-latency interoperability between software-based media functions within processing clusters.
The project was developed by the EBU, CBC/Radio-Canada, and Grass Valley, and it is currently governed by the Linux Foundation.
MXL removes latency bottlenecks by replacing traditional IP transport like ST 2110 with shared memory for internal processing, allowing for more efficient, software-defined live production workflows.
No, Grass Valley positions MXL as a complement to ST 2110. While ST 2110 is used for transport, MXL targets the intra-cluster communication layer to improve efficiency between software functions on the same compute node.
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source
