CODEWITHSUNDEEP
sql-serversql-server-2008elasticsearch
anders
anders

Reputation: 76

How do you keep an ElasticSearch cluster in sync with a DB?

The challenge here is about how to keep a subset of data residing in an Sql Sever 2008 R2 in sync with an ES cluster, with minimal effort and maintainance. I am aware of both changetracking in Sql Server and JDBC River. But both solution will rely on there beeing an Index Manager to extract and transpose the data. So how do you guys do it? Are there any ways of keeping the two in sync using frameworks or any standard applications? It would be preferable if there was a solution was based on .NET, but this is not an "must have".

Edit: This is to be implemented into a system where its not feasible to get hold of the events i would like to manage the index. In other words, no SOA or pub/sub is available. And implementing this is slightly out of scope.

Upvotes: 2

Views: 2105

Answers (1)

gardenunez
gardenunez

Reputation: 41

I would not use any river for this purpose since they are going to be deprecated by elasticsearch team, my suggestion ( depending on your specific use case ) would be to use queue approach for synchronization. Details in here and from there you can adapt to you specific case.

Upvotes: 1

Related Questions