That is correct. We simply built a storage engine technology. You run it yourself on your own servers. Everything is open source except for the core compression algorithms that are loaded as a proprietary dynamic library. Being a young startup, we made the choice not to open source this part for now (see my other comment higher in this thread).