Hi,
I'll take, as an example, an import of a source code tree. This consists of small, highly indexable java files - part of our source tree.
ContentStore: 1018 files @ 9.96MB
Lucene: 6.3MB
MySQL: 13.8MB
This example is skewed towards a larger database and Lucene indexes as the files are small and indexable. If you had larger documents such as Word or PDF documents, then the ratios would be skewed towards a larger content store. The other end of the spectrum would be the storage of large MP3 and other large, non-indexable content.
If you start from a clean install and just copy in (via CIFS, FTP or WebDAV) a tree of documents that represent a typical use-case, then you can look at the sizes of each and get a good idea of how it will grow.
Regards