Re: [Neo4j] Permgen out of memory upgrading to 2.0.3 and SDN 3.0.1

Do you use gremlin? Do you shut dow your test databases cleanly after each test? Sent from mobile device Am 28.05.2014 um 06:40 schrieb Bytor9 : > Sorry to post here. But I refuse to ever use SO. I do NOT, nor will I ever > have an account with them. > > So I recently upgraded our app to u

[Neo4j] jexp batch importer

Hi, would it be possible to add some basic error logging to the jexp batch importer especially for the situation where some records are "skipped" as part of the rels loading but the process itself continues without errors .. sometimes its hard to find the ones skipped in order to resolve the inp

[Neo4j] Permgen out of memory upgrading to 2.0.3 and SDN 3.0.1

Sorry to post here. But I refuse to ever use SO. I do NOT, nor will I ever have an account with them. So I recently upgraded our app to use Neo4J Server 2.0.3 and SDN 3.0.1. We are using Ansible and it is updating a VMWare Fusion VM on my Mac. Now when I run our build I get PermGen Out of Memor

Re: [Neo4j] Using Neo4j embedded in Java applications

6.5M should have no trouble. Sample code? Wes On Tue, May 27, 2014 at 10:36 AM, Qaroui Oussama wrote: > Hello! I'm using neo4j to store more than 6.500.000 nodes, with no > relationships. > When I try to get a specific node using findNodesByLabelAndProperty or by > executing a cypher query in my

[Neo4j] Finding shortest path using the distances which are properties of the relations

Can se be able to find the shortest path using the distances which are properties of the relations? relations here represent the path between two nodes -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving

[Neo4j] Using Neo4j embedded in Java applications

Hello! I'm using neo4j to store more than 6.500.000 nodes, with no relationships. When I try to get a specific node using findNodesByLabelAndProperty or by executing a cypher query in my app I get no results . I created a database with 60.000 and both methods work perfectly . Is it possible that

[Neo4j] Re: Another Native-Java-DSL for the Cypher Language

Hi Wolfgang, So many folks are so busy at the moment and with the U.S. Memorial Day holiday, our thoughts have been elsewhere, but all that said, this effort of yours is AWESOME and welcome -- not only is it a great example of "learn by scratching your itch," but you are doing something that ha

Re: [Neo4j] Does anyone know why Graph Database (Neo4j) consumes more memories(RAM) and has bigger file size(graph.db) than Relational's?

Good point Craig, you touch on an even more fundamental reason why one might see more memory as a good thing: there are many algorithms that are known to be bounded by what is sometimes called the space time complexity trade off. In such cases it is provable, that for given set of hardware the onl

[Neo4j] Closeness Centrality and SingleSourceShortestPathBFS

Neo4j version: 2.0.3 SingleSourceShortestPathBFS is set to CostType=Integer via it implementing SingleSourceShortestPath. If I want to calculate the ClosenessCentrality if have to set it to ClosenessCentrality as well and its CostDivider too. An anonymous implementation would be (as in the test

[Neo4j] Re: Neo4j A Star algorithm performance questions

Ah Craig...two last info.: I was wrong previouslyI'm not sure if node 1 and node 2 are really directly connected...it depends on the OSM file; what I'm sure about is that I reduced to the minimum the complexity of my code by considering only 2 kind of relationships (the ones I put in th

[Neo4j] Re: Neo4j A Star algorithm performance questions

Hi Craig thank you for answering to me Actually I tested the code with this JVM settings: -Xms750m -Xmx4G -XX:+UseConcMarkSweepGC (on my laptop I have 8G of RAM with Ubuntu OS) I used this kind of settings for neo4j: nodestore_mapped_memory_size=250M relationshipstore_mapped_memory_size=3G node

Re: [Neo4j] Re: Neo4j A Star algorithm performance questions

Reading your previous messages I see a comment that "A* from node 1 to node 2: 1416 millis". Are you saying that A* from a node to a directly connected node is taking 1416ms? That does seem insanely long. I could only imagine two possible reasons: - Your JVM is using OS swap space (ie. heap is

Re: [Neo4j] Re: Question about Relationships number and their influence on algorithms

Hi, I just took a peek at the algorthm for the cost at https://github.com/neo4j/neo4j/blob/master/community/graph-algo/src/main/java/org/neo4j/graphalgo/impl/util/GeoEstimateEvaluator.java . I did not double-check the maths, but this does not look like a distance function over a sphere. This look

Re: [Neo4j] Does anyone know why Graph Database (Neo4j) consumes more memories(RAM) and has bigger file size(graph.db) than Relational's?

In my opinion the storage of relationships can (for many data models) take much more space than would occur in an RDBMS. The reason for this is that the relationships contain more information (bi-directional linked lists) allowing fast traversal of the graph in either direction. This is not true of

[Neo4j] Re: Question about Relationships number and their influence on algorithms

Hi Mattias, hi Michael Thank you for answering me @Mattias: I was sure about what you wrote about relationships and influence on algorithms, but now I'm confident that I was right :). But, on the other side, now I'm totally confused on the reason why the AStar algorithm has so poor performance