Clarification regarding loading data into OrientDB

[praveen....@tigeranalytics.com]

I currently have a 3 node OrientDB cluster and i'm planning to store disjoint sets of data in different servers. One main reason for

this is to gain performance improvement during data load and to prevent excessive replication.

My idea is to restrict each cluster of a class to one server alone, based on config such as this : "clusters": { "internal": { }, "index": { }, "client_usa": { "servers" : [ "usa", "europe" ] }, "client_europe": { "servers" : [ "europe" ] } 

based on this link : http://orientdb.com/docs/2.2.x/Distributed-Sharding.html

Will this approach prevent replication, Actually i do not want data to be replicated