Back to the main page
Q&A: Robert Greene on the Midsize Appeal of Oracle NoSQL Database
Robert Greene, product manager at Oracle, has spent years working with the world's most advanced alternative database technologies. We asked him when and why midsize organizations should consider deploying a NoSQL database—a low-cost, highly available, and extremely scalable alternative to relational database technology.
Q. Could you briefly describe the difference between NoSQL and traditional relational databases?
A. NoSQL databases are implementations of distributed key-value stores. They are designed to deal with scale-out of simple data and simple queries. Relational databases such as Oracle Database 12c, are designed to store, manage, and query complex data and complex relationships among that data using what is known as an SQL JOIN operation.
To support JOIN operations, you need all the involved data or its indexes in one place. This presents a problem, usually requiring you to scale vertically, using a more powerful machine for your database. By contrast, the simplicity of a NoSQL database means you can expand horizontally very quickly and at minimal cost simply by adding a number of smaller machines.
Q. What drives the demand for this kind of solution?
A. A big driver is the agility it brings to the line of business by allowing a change to the data model for new feature rollout without having to go through a complex change management process involving a DBA. Also, certain kinds of business data are simple, easy to manage, and don't require all the features and functionality of an enterprise relational database, so a lower-cost NoSQL database makes more sense. Plus, the NoSQL scaleout architecture makes it ideal for supporting the growing demands of big data. Finally, globalization is also a key factor since NoSQL's built in high availability makes it faster and less expensive to expand databases across multiple geographic regions.
Q. Why would a NoSQL database be particularly attractive for a midsize company?
A. Midsize companies are more cost-sensitive, and NoSQL databases offer a highly affordable alternative to relational databases for simpler types of data. Also, midsize companies want to be able to grow quickly, yet incrementally. Using NoSQL, if demand for a new product suddenly takes off, a midsize company can quickly and cost-effectively add more machines to support that growth. NoSQL will automatically take advantage of the added hardware.
Q. What value is Oracle adding with its NoSQL offering?
A. First, Oracle has a long history of success with mission-critical data, and all that experience adds significant value to Oracle NoSQL Database. For example, unlike many competitive products, we deliver powerful support for transactional processing. Second, Oracle's solution supports extremely large-scale server clusters while still ensuring consistent, predictable performance.
Finally, Oracle NoSQL Database integrates with Oracle's software stack and more, including Oracle Event Processing, Oracle Coherence, Oracle Spatial and Graph, external tables, Hadoop, and so on. Integration is vital since no database operates in a vacuum—especially in the era of big data, which requires integration across multiple technologies to deliver its promised value.
Q. What is the relationship between Oracle NoSQL Database and Oracle Berkeley DB?
A. Oracle NoSQL Database is built using the proven Oracle Berkeley DB as the underlying storage engine. Mission-critical applications built on Oracle Berkeley DB already serve millions of users, giving Oracle NoSQL Database an edge when it comes to quality and reliability.
Learn more about Oracle NoSQL Database.