I love Postgres but wouldn't want to use it for everyting.
Taking graph databases, for example, Postgres (via Apache AGE) stores graph data in relational tables with O(log(n)) index lookups and O(k ⋅ log(n)) traversal*
Whereas a true graph database like Neo4J stores graph data in adjacency lists, which means O(1) index lookups and O() traversal.
That's a massive difference in traversal complexity for most graphs.
*k is the degree of the node (number of edges connected to a node).
Taking graph databases, for example, Postgres (via Apache AGE) stores graph data in relational tables with O(log(n)) index lookups and O(k ⋅ log(n)) traversal*
Whereas a true graph database like Neo4J stores graph data in adjacency lists, which means O(1) index lookups and O() traversal.
That's a massive difference in traversal complexity for most graphs.
*k is the degree of the node (number of edges connected to a node).