Skip to Content

MongoDB and Couchbase try to bridge old and new database worlds


Most of the buzz in the database world for the past few years has been around NoSQL (or not-only-SQL) databases, while traditional relational SQL databases were seemingly relegated to also-ran status. So it’s worth noting that at dueling NoSQL database conferences this week, there was a big focus on ensuring that popular NoSQL databases work nicely with products normally seen as part of the older SQL realm.

For the uninitiated, a relational database—like the kind on which Oracle (ORCL) built its multi-billion-dollar business—primarily deals with data that can be stored in rows and columns. A NoSQL database like MongoDB or Couchbase, on the other hand, handles more complex and unruly unstructured data, which does not conform to that common row-and-column format.

NoSQL has garnered a lot of traction with developers because much data generated from modern applications tends to be unstructured. Developers use NoSQL databases to work with such unstructured data as geolocation information and machine log files and to build more sophisticated applications on top of them.

At its conference in New York, MongoDB released a software connector that lets users hook their databases up to the popular business intelligence and visualization services like Tableau (DATA) , IBM (IBM) Cognos BI, SAP (SAP) BusinessObjects, and Qlik that are typically used with SQL databases. SQL, by the way, stands for structured query language.

Companies’ business managers use these analytics tools to get a better view of their data and glean trends. The problem is that these analytic services have traditionally been hardwired to work with relational databases and tools that require the use of the SQL language, which makes them difficult to use with NoSQL databases that are incompatible with SQL.

This connector was the missing piece,” MongoDB vice president of strategy and marketing Kelly Stirman told Fortune in an interview.

Meanwhile, in Santa Clara, Calif., Couchbase showcased a new specialized query language that it says will allow its own NoSQL database to work with Tableau as well.

The new N1QL language is based on SQL and will allow developers to probe their data much like they would if their NoSQL databases were relational databases. The bigger deal for business teams, however, is that this language allows Couchbase’s NoSQL database to work with the familiar data analytics tools they might already be using.

“It makes it easier for Couchbase to plug into the business intelligence reporting items out there,” explained Couchbase CEO Bob Wiederhold. “Surprisingly, we are the only NoSQL database that has a SQL-based query language.”

So why do these two cutting-edge database startups need to work with a language that’s been around for roughly 40 years and is so intertwined in the relational database world dominated by Oracle and SAP (SAP)? Because they can’t ignore how prevalent the language is and if they want their products to gather the type of market share seen at the legacy giants, they need to figure out how to make their own technology compatible with the past.

While no one doubts how important NoSQL is for modern-day application development, the fact that SQL has existed for so long and is so common in the way businesses analyze and process data, it’s simply too big to ignore. And frankly, most of the world’s transactional data, you know, little things like a company’s sales and supply information, is handled by SQL databases.

“[This is] not about the world going against SQL,” said Wiederhold. “It is a transition.”

Business teams at companies want to keep using their preferred data analytics systems like the Tableaus and Informaticas (INFA) of the world. Although NoSQL databases may be hot with coders, it’s not enough to force the business side to have to learn to play with a NoSQL database to retrieve their custom analytics. They just want the tools they currently have to work.

“It is ironic,” said Tableau product manager Jeff Feng who wrote a blog post on the Tableau and MongoDB connector and spoke at Couchbase’s recent conference. “SQL is not required to do this, but you have an ecosystem of technology that has evolved over time around SQL.”

As it stands, the older data analytic and even relational database companies stand to benefit from both of these SQL-centric announcements, because now their products can theoretically work with multiple NoSQL vendors without a lot of tweaking.

Couchbase’s own marketing materials on its new SQL-query language even describe how its product will benefit Oracle developers, so that tells you a little something about how big a piece of the database market Oracle still holds.

But every convert counts, and this week’s NoSQL news shows that the leading startups in the space will be doing all they can to win over new customers.

Subscribe to Data Sheet, Fortune’s daily newsletter on the business of technology.

For more on big data, check out the following Fortune video: