Scalable / Replicated Database

Question

I have a special requirement for storing financial transactions. This requires transaction support, high availability, fault tolerance, and horizontal scalability. I am looking for an open source solution.

Now, as I understand it, none of the free RDBMSs (like Postgres or MySql) provide this. I used cassandra for a previous project that had a lot of convenience in terms of replication and shape. Wow, I like its performance too. But, unfortunately, there is no Cassandra ACID transaction support, as it was built with CAP theory in mind.

Now I'm not even particularly good at excellent / fast performance. I'm looking for performance that is slightly greater than or equal to RDBMS, but with replication and outline. The reason is that we are looking for many servers to handle the load, and each instance must be able to serve some client x at any given time.

Can anyone think of a free (or very cost-effective up to $ 500) RDBMS database or Not in sql with repository - transaction support (commit / rollback) - replication - fragments.

We currently have ravendb in our selection list. Do you think this is suitable?

PS: I looked at the following links. Can you recommend a database that scales horizontally? - discussed in 2008 and very old.

Answer 1

RavenDB will be a good party. It can be scaled horizontally, supporting edging and replication out of the box. It has ACID transactions and an automatic transition to another resource.

See here http://ravendb.net/features for more details.

Answer 2

Take a look at Volt. (Community Version)

Answer 3

You can try scaling using multiple instances / copies of the same database and synchronize them with JTA, which provides distributed ACID, useful for easy ACID scaling. Although records are linearly expensive, reads can be distributed between nodes, and usually reads more often than reads.

This article shows an example of this approach, the code is based on a specific Java ORM tool ( JEPLayer ), but the basic principle of JTA can be applied to various JTA-compatible database tools.