Potential Bug : Double/Float performance issues

[Kamilos]

Hello,

I found something which is very strange.

I have a graph with 100 000 nodes where a node has a property "age".

If i want to update the property "age" of 50 nodes with the Java API 

Node node;

ArrayList<Node> toUpdate=new ArrayList<Node

time1=System.nanoTime();

try (Transaction tx = graphDb.beginTx()) {

for(int i = 1;i<=50;i++){ 

node = graphDb.findNode(labelPerson, "name", i);

if(node != null)

toUpdate.add(node);

}

for(int i = 0;i<toDelete1.size();i++){

toUpdate.get(i).setProperty("age", Math.random());                   //SOLUTION 1  // generate a double between 0.000000 and 1.0000000

toUpdate.get(i).setProperty("age", String.valueOf(Math.random())); //SOLUTION  2 // same number but stored as a string

}

tx.success();

}

time2=System.nanoTime();

If I store the random double value as a double, it takes a lot of time to perform this operation. (201 seconds !!!)

Same issue if it is stored as a float. 

However, If I store the same random value but as a String it is instantaneous.


Any reason why ? Is it related to the internal Lucene Index having issues with this ?

While inserting the data, I created the index the following way :

public static String propertyAge = "age";
inserter = BatchInserters.inserter(new File(Const.DB_PATH_Neo4J));
inserter.createDeferredSchemaIndex(personLabel).on(propertyAge).create();

I using Neo4J 3.0.0

[Michael Hunger]

Can you raise a GitHub issue for this?

--
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/d/optout.