Using neo4j for multiple edges between nodes

[pdj...@gmail.com]

Hi,

I'm quite new to neo4j and graph databases. I'm considering using it for one of my projects where each pair of nodes has multiple edges between them. Each edge has a set of properties which I need to query. If an edge has property X that equals Y(X may not be unique), then the edge is added to the results.

My concern is that this design makes performance degrade because between two nodes, there's no index and lookup time approaches the number of edges between the nodes. Is there another design that optimizes performance ?

Thanks

[Peter Neubauer]

Do you have any size estimations on the number of relationships between two nodes, and the total number of nodes in the graph?

/peter

G:  neubauer.peter
S:  peter.neubauer
P:  +46 704 106975
L:   http://www.linkedin.com/in/neubauer
T:   @peterneubauer

The graphgist challenge - Bring your own graph - http://www.neo4j.org/learn/graphgist_challenge

--
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+un...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[pdj...@gmail.com]

Eventually the number of nodes should reach 10^5. Number of relationships between two nodes shouldn't be more than 100.

[Michael Hunger]

That shouldn't be a problem at all.

Michael

[pdj...@gmail.com]

So, neo4j can handle 5*10^11((10^5)C2 *100) entities of data without performance degradation ?

[Michael Hunger]

How do you get from

10^5 = 100k x 100 = 10^7 to 10^11 ?

[pdj...@gmail.com]

10^5 nodes. For worst case scenario, assume each node is connected to every other node by atleast one edge. That means we'll have a minimum of (10^5)C2 edges(~5*10^9). Now add the fact that there are 100 edges between two node and you get 5*10^11 edges. 

[pdj...@gmail.com]

On an unrelated note, I was wondering if instead of having multiple edges, I have a single edge and store a 2D-array of the data, can I hash/index this 2D structure to improve lookup time during queries?