Oracle Berkeley DB vs. Oracle NoSQL Database

Oracle NoSQL Database is well-suited for you if your data formats are not consistent, if you have limited hardware resources, if you higher data throughput (whether the database is on the cloud or running locally), and if you don't need a declarative query language to maintain a standardized schema of your data. If you need reduced data redundancy and require ACID compliance, you are better off finding an SQL database solution.

• Data-model flexibility. Unlike RDBMS solutions, Oracle NoSQL does not restrict you to a predefined set of data types.
• Ability to Handle an Increased Amount of Traffic. As Oracle NoSQL can process queries much quicker than Oracle Database, Oracle NoSQL is able to respond to a lot more queries in the same amount of time.
• Data-model simplicity. In SQL-oriented databases, there is a learning curve in learning the relationship between databases, tables, rows, and keys. On the other hand, Oracle NoSQL's key-value based storage is much easier to get the hang of.

• Fewer analytical functions to choose from. When compared to Oracle Database, there is significant difference in the amount of built-in analytical functions.
• Eventual data consistency. It is not guaranteed that a write or delete query will be immediately visible for subsequent queries.
• Data redundancy. As there are no mechanisms that insure data integrity, users are more likely to have redundant data across their documents.

I have not used any other types of NoSQL databases.

• We pay less for computing resources, as Oracle NoSQL databases respond quicker than our previous SQL databases.
• Our database administrators and software developers do not need to worry about "data massaging" and can focus on perfecting application logic.
• Oracle NoSQL has built-in integration to other Oracle products, so we didn't not need to spend money on building custom integrators or higher additional developers.