RenderStream Networking

Disguise RenderStream is a protocol and technology developed to enable real-time streaming of rendering data between systems, such as media servers and rendering engines.

It is designed for synchronised, high-quality, and low-latency workflows, commonly used in applications like:

• Virtual production
• Live events
• Immersive experiences

Bandwidth Considerations

• RenderStream handles high-resolution, high-frame-rate content, including textures, video, and rendering data.
• Uncompressed data significantly increases bandwidth demands.

Transport types

Streams in RenderStream can either be compressed or uncompressed.

Uncompressed streams

RenderStream Uncompressed Streams (sometimes shortened to RSUC) are low latency, high-quality, and require much higher bandwidth.

Uncompressed streams require specific hardware:

• Network Interface Cards (NICs): 25Gb or 100Gb Mellanox network interface.
• Multiple Machines: dedicated render nodes render and send content via uncompressed streams to actor and director machines.
• Genlock: required when rendering multiple fragments that are output by multiple actor machines.

Supported uncompressed formats:

• RGB 4:4:4 8bit
• RGB 4:4:4 10bit
• RGB 4:4:4 12bit
• YUV 4:2:2 8bit
• YUV 4:2:2 10bit
• YUV 4:2:2 12bit

The transport format can be configured in the cluster assigner settings of the workload: RenderStream Layer

Compressed streams

Compressed streams use H.265 compression to reduce bandwidth needs while providing consistent content quality. Compressed streams use an order of magnitude less bandwidth than uncompressed and can be used on the 1G, 10G, or 25G+ networks. A limited number of streams can be encoded and decoded by the hardware on the cards.

Supported compressed formats:

• YUV 4:2:0 8bit

The transport format can be configured in the cluster assigner settings of the workload: RenderStream Layer

RenderStream Network Configurations

RenderStream remote mode requires a high-bandwidth network to connect the dedicated render node(s) to the VX and/or GX range media server(s) being used to composite and output the content.

We have standardised on Mellanox ConnectX-5 (25GbE) or ConnectX-6 (100GbE) network interface cards (NICs) on our products. This is because they support Rivermax, which we use to enable RenderStream Uncompressed.

Fabric is a preconfigured network switch that works with RenderStream or ST 2110 video if using the IP-VFC and is sold by Disguise. Fabric is recommended as a known-good solution to work with RenderStream.

We recognise that users may already have access to a high-speed network switch, or have requirements to accessing a network switch with their own configuration. The following minimum requirements must be met for any network switch being used alongside RenderStream:

• The network switch must be able to support the bandwidth being used by the NICs on the render node(s) and the VX/GX range media servers being used. For example, if the NICs are 100GbE then the network switch must be able to deliver up to 100GbE per port or the use of breakout cables (i.e. 4x 100GbE). If the NICs are 25GbE then the network switch must be able to support 25GbE per port or the use of breakout cables (i.e. 4x 25GbE)
• IGMP snooping must be enabled on the required network ports used to connect to each render node and VX/GX range media server being used in order to manage and control the flow of Layer 3 multicast traffic.
• Jumbo packets should be disabled on the required network ports used to connect to each render node and VX/GX range media server being used.
• RDMA over Converged Ethernet (RoCE) version 2 must be supported, specifically lossless mode.
• Quality of Service (QoS) must be maintained to prioritise media streams over control traffic.

Known-good SFP and Fibre for RenderStream

Below is a table of recommended SFP types to use with RenderStream.

Mellanox NIC port speed	SFP Connector Type	Validated SFPs
25GbE	SFP28	Mellanox MMA2P00-AS FS SFP28-25GSR-85
100GbE	QSFP28	Mellanox MMA1B00-C100D FS 71009

We strongly recommend the usage of OM4 fibre, which is a high-bandwidth multimode optical fibre.

LC to LC OM4 fibre cables should be used for 25GbE networking and MTP to MTP OM4 fibre cables should be used for 100GbE networking.

MTP to Quad LC OM4 breakout fibre cables should be used for connections between 100GbE and 4x 25GbE.

Additional Considerations

NIC Throughput

Even with 100Gb Ports, actual throughput is typically around 80Gb due to protocol overheads, error checking, and operational processes.

Using the Assigner and predictions on workload traffic you can aim to calculate the expected throughput needed per workload.

Cluster Rendering

Cluster Rendering increases cluster management network traffic, which at very large scales can become a bottleneck. Please refer to the RenderStream Network Configurations section.