Broadcasters and product developers gain new flexibility in their ST 2110 IP workflows with Macnica's MEP100 SmartNIC. The card now supports both 25GbE and 100GbE media flows, along with improved GPU processing, Windows-based application support, and real-time metadata capabilities.
Enhanced Flexibility for Modern Workflows
This enhancement allows users to scale their SMPTE ST 2110 workflows based on channel density, infrastructure needs, and specific application requirements. Andrew Starks, Director of Product Marketing at Macnica, highlighted the significance: “Never before has this kind of flexibility been offered in a SmartNIC. Our customers can now align network bandwidth with their application-specific channel requirements without being locked into a single network speed.”
Adapting to Changing Workflows
The addition of dual-speed support reflects the evolving nature of broadcast and media workflows. Developers can optimize network bandwidth for various applications, from graphics generation and video server playout at 25GbE to high-density, uncompressed workflows for switching and multi-channel processing at 100GbE.
Improved GPU Integration
The MEP100 introduces enhanced support for GPU-accelerated workflows through GPUDirect, which facilitates efficient media stream transfer into GPU memory. This improvement is particularly beneficial for Windows platforms, where new DirectShow filter support simplifies integration with applications like vMix.
Metadata and Emerging Workflows
The platform also supports SMPTE ST 2110-41 Fast Metadata, enabling the transport of time-aligned metadata alongside video and audio streams. This capability is crucial for emerging workflows using Apple ProRes 422 within ST 2110-22 environments.
Hardware-Accelerated Processing
Built on Macnica’s Altera FPGA-based architecture, the MEP100 offers hardware-accelerated processing of ST 2110 streams. This includes support for seamless protection switching, PTP synchronization, GPUDirect-based data transfer, and kernel bypass to reduce CPU load while maintaining deterministic, real-time performance.