Simplifying the Deployment of Hadoop Applications

Michael Vizard
Sep. 11 2013, 10:00AM EDT

While there’s generally a lot of excitement these days for all things Hadoop the simple fact is that there’s not a whole lot of expertise out there in terms of IT people that know how to build, roll out and then manage an instance of Hadoop. That creates a frustrating conundrum for developers looking to build Big Data applications on top of Hadoop at a time when interest in those applications is staring to reach a frenzy in the enterprise.

To help solve that specific issue Continuuity has created an application server platform for Hadoop that allows Java developers to deploy Big Data application on an instance of Hadoop running in the cloud or on premise in a matter of minutes.

Img Credit: Continuuity

According to Continuuity CEO Jonathan Gray, Continuuity has come up with a set of Java and REST APIs that allows developers to invoke a Reactor engine to automate the deployment of their applications on top of Hadoop, Hbase and the rest of the components that make up a Hadoop stack.

Instead of waiting six months or more for an internal IT organization to figure out how to deploy Hadoop, Gray says Reactor gives developers the option of buying a Hadoop stack in the cloud that allows them to instantly deploying their applications. Once deployed than developers can figure out how to integrate something that provides business value with the rest of the legacy applications that make up the average enterprise.

Interest in Hadoop applications in the enterprise is high, but progress has been slow. Most enterprise IT organizations are still piloting Hadoop applications with a modest amount of data. That slow pace creates tension with line of business executives that are looking to tap into Big Data today, which often means they are turning to software-as-a-service (SaaS) applications rather than wait for developers and internal IT organizations to master all the intricacies of Hadoop.

Much of this is simply comes down to a classic build versus buy debate in the enterprise. But for developers that want to stay relevant in the age of Big Data, they clearly need to deliver value from Hadoop investments sooner than later. Where those applications run when they fully mature may be open to debate. But right now the line of business executives that fund these projects are not generally known for their patience, which means most developers are being asked to deliver something of value to the business much sooner than actually later.

Michael Vizard



  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.