Cypher, being the "SQL for neo4J", can be used to update nodes. When using any relational DB, even a crappy one like MySQL, I have never seen a problem when I wanted to update each and every row in a table with one query, even when the table contains millions of rows.
(This is jsut an example, the same goes for creating a million links between nodes, based on a query that selects nodes of a different type where some properties match.) So, I ask myself, why does Neo4J go out-of-memory when I want to do something like that? It seems as if it tries to load every node into memory first before doing something, but that can never be a future-proof solution, since you can't expect to always have enough memory avaliable to store your whole database. So I am wondering, what can I expect in future versions of Neo4j. Is this something that will be solved (so that I can run and manipulate a somewhat decent test database on my laptop with limited memory, like i can with a Postgres or MySQL DB). (I am not really interested in all kinds of 'workarounds' that people may have found in order to circumvent these limitations). -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
