Graph Databases like most other NoSQL database don't impose a schema. That
helps most people to write more flexible applications in a faster way as you
minimize roundtrips between schema migration and application development.
That said, creating a graph model (e.g. on a whiteboard or with a diagramming
tool) helps a lot figuring out how to support the use-cases that you want to
take care of.
After having developed your model, you import data into the graph databas using
that model and then querying according to the model.
There are some additional pieces in Neo4j that help you manage your data
structures:
- Labels on nodes denote roles that entity in your domain play (like :Product
:Person:Father:Employee:Developer :Location:Venue:ConferenceCenter etc.)
- Relationship-type declare the types of connections you have between enties,
e.g. (:Person:Actor)-[:ACTED_IN]->(:Movie) or
(:User)<-[:ISSUED]-(:Order)-[:CONTAINS {quantity,price}]->(:Product)
- Indexes on :Label(property) and unique constraints on :Label(key-property)
Here are some things that might help you get going:
The free graph databases book: http://graphdatabases.com
Ian Robinsons Presentation at QCon London: New Opportunities for Connected Data
https://vimeo.com/89075703
The modeling page on neo4j.org http://www.neo4j.org/develop/modeling
and last but not least, live, interactive data modeling examples:
http://blog.neo4j.org/2014/02/graph-gist-winter-challenge-winners.html
If you still have questions, please don't hesitate to ask.
For running independent instances, you have some options
- put all in one graph, oftentimes your domains are actually connected (size is
not really a limit)
- change the database location property in your conf/neo4j-server.properties
config file according to which database you want to use
- install multiple servers with multiple data directories and http ports
configured
- use Neo4j in embedded mode which allows you to freely choose where the
database is located
There are some really cool community tools, like Jim Salmons CPanel
http://jim-salmons.github.io/neo4jcp/
Or graph application frameworks like structr that take care of that.
Cheers,
Michael
Am 17.03.2014 um 23:42 schrieb Amit <[email protected]>:
> Hi,
> I am coming from Oracle background and wanted to know if there any anything
> equivalent to different schema of a oracle db in Neo4j.
>
> We can have multiple db instances running on a machine and in each instance
> we can have different schema/users, these schema can have multiple tables.
> This is logically grouping of data into a schema for tables which are related
> to a application.
>
> I understand this is totally different from a Neo4j use case but if i have
> unrelated data how do i store them in Neo4j ? how to i create different
> instance of Neo4j.
>
> E.g. - In Cassandra we have Keyspaces having tables / column families..
>
> - Amit
>
> --
> 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.
--
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.
