Mesh (full-mesh, peer-to-peer) is the simplest WebRTC conferencing topology: every participant sends their media directly to every other participant, with no media server in between. For two to a few people it is ideal — lowest possible latency, no server cost, and naturally encrypted end to end. The problem is that it doesn't scale: each participant must encode and upload a separate copy of their stream to everyone else, so upload bandwidth and CPU grow with the number of participants and become impractical past roughly four people. Beyond that, conferences move to an SFU, which lets each client upload just once. Mesh remains the right choice for one-to-one calls and tiny groups.
The book · Volume 5 of 7
Audio for Video: A Complete Guide to Sound in Video Products: From Jitter Buffers and Lip Sync to Atmos, Quality Metrics, and Neural Codecs
Mesh audio is perfect for two or three people and falls apart once each client must upload a stream to everyone else. This volume covers mesh topology in detail next to SFU and MCU, with bitrate decisions, jitter buffers and loss recovery, so you know when to stay peer-to-peer and when to add a media server. Written by Nikolay Sapunov, CEO at Fora Soft.
Your peer-to-peer mesh calls break down once more than a few people join?
Tell us your typical call size and networks and we will recommend when to move from mesh to an SFU, with cost and latency estimates. 250+ video products shipped since 2005.
