Any chance to share your db with us privately if you want to? Will try to find some time to look into it.
Cheers, Michael ---- (michael}-[:SUPPORTS]->(YOU)-[:USE]->(Neo4j) Learn Online, Offline or Read a Book (in Deutsch) We're trading T-shirts for cool GraphGist Models Am 07.03.2014 um 19:33 schrieb Ziv Unger <ziv.un...@gmail.com>: > 1. According to Michael Hunger at neo4j, yes. I've also seen evidence of this > when optimising my own queries on his advice. > > 2. You are correct here, sorry. Will teach me to reply before coffee in the > morning. > > 3. Not sure, someone with better knowledge of the inner workings of the db > would have to weigh in. You could use a profiling tool like > https://github.com/mneedham/cypher-query-tuning, but if your queries don't > return at all, it won't help. I did find that when requesting huge numbers of > results, queries slow down dramatically. If you add a limit 100, does it > still not return? > > I had a situation where even with a completely tuned query, returning > thousands of results still took ages. If the limit helps, perhaps paging is > your answer. > > Otherwise, hopefully one of the neo4j guys will weigh in soon. > > > On Friday, 7 March 2014, Zeeshan Arif <arifzees...@gmail.com> wrote: > Thanks a lot for providing some inputs. However, for me, with the initial > posting of records i.e. 3814 relations for primary and some of the secondary > have 40K+ relationships, it is still not performing well - infact doesn't > return. > > Few questions: > > 1. Are you sure that with one kind of relationship between nodes, skipping > the label helps? Intuitively, qualifying everything should always help.. > isn't it? > > 2. C-->E will look for any length path? I thought it will look only for > direct neighbors by default and only if I mention *1..x, it will go x deep. > > 3. Any idea if UNION is expected to work better than OPTIONAL MATCH clause > and combining results together? > > Regards, > > - Zeeshan > > > > On Thursday, March 6, 2014 11:15:42 PM UTC-6, Ziv Unger wrote: > As an added note, if you DID want any length of path between (a) and (c), you > can just use [r3:knows*1..x] where x is the upper limit of the path length. > If you have a lot of nodes, I would personally limit the upper range for > performance. > > -- > You received this message because you are subscribed to a topic in the Google > Groups "Neo4j" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/neo4j/5hapONQs7Qo/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > neo4j+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > > -- > 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 neo4j+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. -- 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 neo4j+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
