Hello, I encountered a problem when using livekit, the problem is the following, we periodically call the whole team in one session (about 50 people), while the livekit-server is located in a private cloud, when all 50 people are in different places, the meeting goes well, however, there are situations when 30 people are in one building, and someone is sharing the screen, in this case the incoming channel is very heavily loaded and crashes begin. That is, we have a fairly productive local network, and the incoming channel is approximately 100 mbit/s and the entire channel is clogged with 30 takes of the demo track. As I understand it, the problem can be solved by installing an additional relay server in the building. I read on the blog (https://blog.livekit.io/scaling-webrtc-with-distributed-mesh/) that they did this for livekit cloud. But even if you use it, it will not solve the problem of the incoming channel in the building because... you need a relay in the building. Is there some kind of open source implementation of relay? Or are there any other ways to solve this problem? Thank you in advance.