SQL Server is marketed as a relational database. But is it? Is the idea that SQL Server is a relational database just another great myth like the carrots and WWII?
At it's core SQL Server is a collection of indexes* and query execution plans. What makes it a relational database is the addition of Primary Keys, Foreign Keys, and that fact that most installations use SQL Server to implement a relational database design. But there are other types of databases that could be built using SQL Server's most excellent indexes and query execution plans.
A friend of mine recently challenged me saying that TejonDB isn't really a graph database. It's really a relational database with a graph layer on top, so that's got to be slower than a database built to be a graph database.
I replied that any graph database is nothing more than an indexed data structure and code to read and write to that data structure. SQL Server is really nothing more than indexes and query execution plans. SQL Server's query optimizer produces query execution plans that are exceptionally fast. Given a clever set of indexes and good SQL code that becomes a great query execution plan, there's no reason why SQL Server couldn't be the platform of a graph database that runs faster than a hand-coded graph database.
That's what TejonDB is - an Object-Oriented / Relational /Graph database built inside SQL Server using clever set of indexes and SQL code that leverages those indexes so that SQL Server's Query Optimizer can build screamingly fast query execution plans. We at Tejon Tech can focus on the data features you need while we rely on Microsoft to deliver the enterprise grade high-availability, security, and reliability. And that's no myth.
* To optimize SQL Server you need to take the view that there is no such thing as a table inside SQL Server. It's all just indexes and query execution plans. A table isn't real, it's just an abstraction of a clustered index.