At the Fluent Conference this week, John Furrier, host of theCube, invited Gary Nakamura, CEO of Concurrent Inc. Startup, to talk in-depth about the trajectory of his company from the seedfunding days to the time when thousands of data driven businesses (including Twitter, eBay, The Climate Corp and Etsy) rely on Cascading as their default framework for building and deploying large scale data processing applications.

As Furrier noted, Big Data developers are focused nowadays on analytics, service layers, cloud and mobile integration.

Concurrent was founded in 2008 by Chris Wensel, a core contributor to a framework called Cascading, a very popular framework for building applications on top of Hadoop. After obtaining some seedfunding, they boosted engineering efforts and focused on adding additional value to the users and deployed applications.

The open source product is a development framework. Cascading makes it easy for developers and people who know SQL, or any data scientist, to deploy an application on Hadoop in a very short amount of time. The framework is free, but the company manages to commercialize some of its value.

Big Data = Big business problems

Nakamura believes the business problem that Big Data created was very disruptive. Hadoop offered a very economical way to store, scale and access that data, but building applications on top of Hadoop was very difficult. Chris Wensel managed to turn this into an opportunity, creating the framework to have a separate business logic layer from the data layer. Cascading is a very logical Java application framework that allows developers to create rich data analytics and data management applications quickly and easily.

Cascading fits into any Enterprise Java development environment. The requirements for the enterprise are not to develop new skills to adopt Hadoop, but to leverage existing skills, systems and investments in their infrastructure.

JavaScript is maturing

The protocol of REST API is changing the game for developers. How is the Concurrent platform working with the whole API construct? In the JavaScript world, they are leveraging on both the backend and front end UI for commercial monitoring products.

There are a couple of different sets of APIs: to build applications there is a Java API, an SQL interface and a machine-learning interface. Developers are offered Cascading SDK. It comes with the framework, plus a set of tools required to build applications.

Developing applications at the lower level API on Hadoop has been very difficult and it required high level skills. Cascading intends to make it easier for mainstream enterprise to build applications. As for the use of SQL, Nakamura has a very simple explanation: SQL is 10x Java, and it represents a must-have for the development world. If Hadoop crosses over into the mainstream skills set, SQL is a prerequisite.