Neo4j replication by filter

Question

I have big database of geolocation and I want to create 100 replication of the databases based on countries i.e. World is my master database and every slave will be having data of one country. Is it possible to do in Neo4j as all documents out there explain concept of HA but how we can slice the database while creating slaves it isn't mentioned anywhere.

And doing so will it affect my db performance?

It will be then individual databases which are not in a cluster. But you can still transport information between them e.g. with a message bus — Michael Hunger, Aug 25 '15 at 11:46
Why do you want to do that in the first place? How big is your current database? — Michael Hunger, Aug 25 '15 at 11:47

score 1 · Answer 1 · edited May 23 '17 at 12:14

1

If I understand correctly, what you want to do is database sharding. Which is not supported in Neo4j, because it's "hard" problem in graphs to find weak spots.

Here great post about it - https://stackoverflow.com/a/21566766/69684

Still you can do that, but information about where data is for each country, you need to store in your application.

Could you please share your current model and tell us more what you want to achieve?

edited May 23 '17 at 12:14

Community

1
1

answered Aug 25 '15 at 09:55

MicTech

42,457
14
62
79

My Database is realtime there will be 100's of update every minutes to be exactly describing my schema it is transportation app which will store current status and changes for every vehicle. (Vehicle)-[:Starts,:Vehicle_Rel,:Route/Country]-(station)-[:Vehicle_Rel,:Route/Country]-(station).....-[:Ends,:Vehicle_Rel,:Route/Country]-(Station) Here there are multiple relations between every nodes that can be 'N'(depending on No of vehicles). So I want to shard my database according to countries/routes so that I can easily get the data instead of querying whole graph to find one vehicle.Any sugg.? – DonRaHulk Aug 27 '15 at 09:02

Neo4j replication by filter

1 Answers1