I am considering Microstream for a project but before adopting it would like to better understand how the framework actually work and in particular what the overhead (indexing, any wrapper objects etc) is per persisted object and per root. I have looked around quite a lot on both the microstream web site and Internet in general but failed to find information on this detailed level....
Any suggestions, short of actually reverse engineer the source code that I see as a last resort would be much appreciated!
Microstream is open-source. You can check the Github repository. There are also many videos on youtube from Microstream Team. Microstream Team also is very active in Java User groups, has many live presentations about this framework, you can participate in these meetings. Microstream JavaDoc is part of the Maven Artifacts in the maven repository.