You can model your data in Oracle NoSQL Database by using Tables, JSON schemas or a raw key-value interface.

Tables are the easiest way to model data. They provide the highest level of abstraction, they are simple to model and should be familiar to any developer. This model also supports secondary indices and table evolution. For more information on the tables API, see Oracle NoSQL Database Getting Started with the Table API.

You can use JSON to model data for your JSON centric applications. If secondary indices or strongly typed keys are not a priority and if keys are going to be modeled manually, then this is a good choice. For more information on JSON, see the Oracle NoSQL Database Administrator's Guide.

Finally, if you want to seralize the data, manage the key structure, manage secondary indices through index views, manage evolution and security through your client code, then you can use the raw key-value interface. This is a good choice if you are willing to write code on to your store to manage things that are either available in the JSON or table interfaces. For more information on the key-value API, see Oracle NoSQL Database Getting Started with the Key/Value API.