There's not too much deep information about them on the web.

How are they implemented and how they work under the hood? I'd like to know that they cause any performance overhead or not, especially in the case of multiple subroutines to decide what's the best practice to create different versions of shaders: subroutines or separate shaders.

Any info or advice?

Have a look at this article: http://www.packtpub.com/article/opengl-glsl-4-using-subroutines-select-shader-functionality

In general, using subroutines should perform better than shader swapping.