Automating the Intelligence of Business Intelligence
Since last summer, Couchbase has been working to marry a derivative of SQL to a document database based on JavaScript Object Notation (JSON). This week, Couchbase released a 4.1 update to Couchbase Server that adds support for Create, Read, Update, Delete (CRUD) operations within the N1QL implementation of SQL as well as covering indexes and prepared statements to reduce latency of complex queries.
While document databases have proven to be a popular alternative to traditional relational databases for a broad range of applications, Ravi Mayuram, senior vice president of products and engineering for Couchbase, notes that SQL is still the lingua franca for querying data stored in any type of database. To accommodate that fact, Mayuram says a document database coupled with N1QL enables IT organizations using just about any standard business intelligence application to take advantage of SQL derivatives to query data without having to get locked into a particular database schema.
In general, the single biggest limitation with database schemas is their rigidity. Every time the data model an organization has adopted needs to change, the IT organization has to modify the database schema. That process is not only time consuming, it requires a lot of relational database expertise. Mayuram says that as a class of so-called “NoSQL” databases, a document database can not only scale out as the data model grows, there is no need to create a schema for any specific data model.
Developed in collaboration with the University of California, San Diego, Mayuram says the intention is to propose the core technology used to create N1QL as a standard extension of traditional SQL.
In the meantime, IT organizations need to come to terms with the fact that the database landscape is now more diverse than ever. While that may be more costly to manage, the ability to more dynamically interact with data to make the business as a whole more agile more than offsets those additional costs.
