It's going to be slow for global algorithms in large graphs, and fast for local traversals in any graph. Which is the best you can hope for, and they're probab

It's going to be slow for global algorithms in large graphs, and fast for local traversals in any graph. Which is the best you can hope for, and they're probably close to the state of the art at doing both in one DB.