SQL is the Rosetta Stone of Big Data, Says Bradford Stephens of Drawn to Scale

Bradford Stephens

Bradford Stephens, CEO and Founder of Drawn to Scale, discussed database trends in Big Data and the concept of real-time and how it varies from one provider to another, with theCube hosts Dave Vellante and David Floyer of Wikibon, live from the Strata 2013 conference.

Talking about the evolution of the field, Stephens mentioned that he has been one of the earliest Hadoop practitioners. He got involved in this market “since we didn’t really know what Big Data was about.” The energy coming from the conference, the amount of cash poured into this field is impressive, he added, reminiscing over the cloud’s early days.

Talking about Drawn to Scale and its play for Big Data, Stephens mentioned always being fascinated by “crazy, complex stuff,” nothing being more complex that a database. Databases are traditionally the cause of all evil, he explained, as when something breaks, when something cannot be implemented, that is where the blame would go. Additionally, when too much data is stored into a traditional database, everything falls apart. This is where Drawn to Scale comes in, as Spire is an SQL database for large, user-centric applications built on Hadoop,

SQL is currently “the Rosetta Stone for data,” Bradford Stephens said, every player trying to integrate SQL queries into Hadoop. As there is a lot of value in this particular integration, to tap into it, a company needs to build the platform from the ground up.

Commenting on the converging industries of MySQL and NoSQL and their interaction with Big Data, he explained that NoSQL lacks features, and developers fail to understand it. An average developer cannot write code against it. On the other hand, MySQL clusters as a solution make room for a business in itself when it comes with operating such clusters with large amounts of data.

Customers that come to Spire from MySQL clusters do so because the demands of the distributive world cannot be met by traditional databases, which include MySQL.

Exploring use cases for Spire, Stephens mentioned three big directions – web and mobile apps, real-time ad serving and telecom applications. The database provided by Drawn to Scale is easily scalable, developers know how to use it and it’s a real-time solution, allowing the creation apps on top of the solution, instead of forcing developers to become database experts. Exemplifying the real-time ad server, he gives the example of a person buying an iPhone and once the payment is processed, getting a tip that there is a store nearby that sells accessories for the new acquisition.

Telecom is a rich market for Drawn to Scale because these customers traditionally have Big Data needs and are now moving from preferred solutions such as Oracle.  Telecom customers want one database that scales to hold all data in one place and allows for diagnose, analysis and generate insights and predictions. Spire also allows them to update all the data they gather in real time, see real time usage and build applications on top of the solution that generates business decisions.

Highlighting the real-time concept, Stephens said his company defines it as a few hundred milliseconds, or how long it takes for a user to leave the website, or as long as it takes for someone to figure something is not working. While they think in terms of milliseconds, other players, including Cloudera’s Impala, define real-time as a few seconds to a few minutes.