> On that page they mention that FlockDb is built on top of Gizzard. I'm > having trouble figuring out just what FlockDb really adds. >
Briefly I *think* it does away with the (traditional SQL) need of one join table per possible relation. E.g. in SQL you normally need one table to map Object A to Object B and another for Object A to Object C. FlockDB adds a way of uniquely identifying each row in the database (a global id), so you can have one 'relationship' table that can represent relationships between any 'global' object. This makes the schema smaller, more flexible and can increase performance in some cases (instead of querying many join tables, I can just query one). Its exactly the custom 'graph like' DB structure I have been using for a while. It works 'quite well' but a true graph db would be more flexible, but maybe not as fast (depending on application and traversal depth) Al _______________________________________________ Neo mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
