How to explain the performance of the Cypher LOAD CSV proposal?

Question

I am using Cypher syntax LOAD CSVin Neo4J 2.1.2. So far, this has been a big improvement over the more manual ETL process required in previous versions. But I encounter some kind of behavior in one case, which is not what I expect, and I wonder if I am missing something.

The cypher request used is as follows:

USING PERIODIC COMMIT 500
LOAD CSV FROM 'file:///Users/James/Desktop/import/dependency_sets_short.csv' AS row
MATCH (s:Sense {uid: toInt(row[4])})
MERGE (ds:DependencySet {label: row[2]}) ON CREATE SET ds.optional=(row[3] = 't')
CREATE (s)-[:has]->(ds)

Here are a couple of CSV lines:

227303,1,TO-PURPOSE-NOMINAL,t,73830
334471,1,AT-LOCATION,t,92048
334470,1,AT-TIME,t,92048
334469,1,ON-LOCATION,t,92048
227302,1,TO-PURPOSE-INFINITIVE,t,73830
116008,1,TO-LOCATION,t,68204
116007,1,IN-LOCATION,t,68204
227301,1,TO-LOCATION,t,73830
334468,1,ON-DATE,t,92048
116006,1,AT-LOCATION,t,68204
334467,1,WITH-ASSOCIATE,t,92048

, Sense node ( ) ID, . , DependencySet node, , . , has Sense node DependencySet node. , , . , CSV .

CSV Lines       Time (msec)
------------------------------
500             480
1000            717
2000            1110
5000            1521
10000           2111
50000           4794
100000          5907
200000          12302
300000          35494
400000          Java heap space error

, , 500 , manual, :

, - 300k 400k , Java- . , , 400 . . 5-7 , . , 300 000 , , " " ? , Neo4J , , , .

, , Sense.uid DependencySet.label , . :

Indexes
  ON :DependencySet(label) ONLINE (for uniqueness constraint)
  ON :Sense(uid)           ONLINE (for uniqueness constraint)

.

: , -, MATCH / CREATE. 3 5 Cypher, .

Answer 1

, Sense LOAD CSV. , , Sense DependencySet Sense node CREATE (s)-[:HAS]->(ds), .

, . , Sense . , . , 400k , . .

LOAD CSV . . : http://jexp.de/blog/2014/06/load-csv-into-neo4j-quickly-and-successfully/

. .

Answer 2

,

MATCH (s:Sense {uid: toInt(row[4])})

. - 200 000 x , Sense , . , - ( , ).

, , .

java Kenny