Our company is using EMQX, with tens of millions of devices connected to the broker. Therefore, we need to have a comprehensive understanding of the operation mechanism of EMQX. Unfortunately, I am currently finding it difficult to locate similar documentation.

Similar to documentation like Kafka:

https://kafka.apache.org/documentation/#design

https://kafka.apache.org/documentation/#implementation

We are currently using the open-source version of EMQX and have encountered some performance issues. Besides seeking help from the community, we also hope to gain a deeper understanding of what performance tuning can be done.

I am currently compiling EMQX on a local Ubuntu system and have been debugging the source code. I have encountered the following issues:

1、After modifying the source code, running the make command is very slow and takes a few minutes to complete the code compilation. Is there a faster method?

git clone https://github.com/emqx/emqx-rel.git
cd emqx-rel
make
_build/emqx/rel/emqx/bin/emqx console

2、It is difficult to trace the code execution path while reading the source code. I think this might be related to Erlang's message model, where a single request requires the cooperation of many processes.

For example, I printed a log in the terminate function, but I don't know which process is sending the message, making it difficult to trace the source of the code