In theory, a graph database is a relational database + recursion. In practice, some relational databases support recursion but SQL is not designed for it and the facility is both limited and often very awkward for many elementary graph operations. Nothing about the design and structure of SQL is optimized for this type of (ab)use case.

Consequently, managing a large-scale, complex graph data model in SQL ends up being an unmaintainable mess and very difficult to optimize for performance. SQL is a select-centric query language, whereas graph data models strongly recommend a query language that is join-centric.