Hi,

maybe I'm missing smth but the very idea of importing database to Neo4j via 
pack of Cypher "CREATE" statements seems to be a counter-productive idea 
for me.

I'd suggest to dump the base into 2 .csv files:

   1. first one describes nodes (and enumerates them via some *my_node_id* 
   integer attribute), columns: *my_node_id,label,node_prop_01,node_prop_ZZ*
   2. second which describes relations, columns: *source_my_node_id, 
   dest_my_node_id,rel_type,rel_prop_01,...,rel_prop_NN*
   
Than import first file via LOAD CSV, than the second one in the same way. 
Don't forget to *add your needed indexes and/or constraints before starting 
import*! This will save you much time and effort.

Mixing creation of nodes and relationships in a single LOAD CSV statement 
turned out to be a risky idea, at least as of early 2.1.* releases. Also 
I'd recomment to import no more than 10000-30000 lines in a single LOAD CSV 
statement - depends on your hardware and especially RAM setup. Just split 
your BIG .csv into parts with appropriate number of lines - that's what I 
did i.e. with my experimental 9 million lines .csv

WBR,
Andrii

-- 
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 [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to