Hadoop gets real

Robust data processing and storage capabilities make Hadoop both wildly popular and wildly complex. Here's how four IT leaders managed to bring Hadoop systems from the sandbox into production.

1 2 3 4 Page 2
Page 2 of 4

Taking Hadoop to the Next Level

Inovalon is another company that's fast discovering how easy-to-use software can drive greater enterprise adoption of Hadoop. A Bowie, Md.-based provider of healthcare data analytics solutions, Inovalon manages data on more than 140 million patients, 220,000 clinical groups and 540,000 physicians. In order to process this information, the company could have chosen a system from one of today's top Hadoop distribution providers, including Hortonworks, Cloudera and MapR. Instead, Inovalon selected Pivotal software, which not only crunches data, but also extracts intelligent insights from big data streams.

Inovalon spent months running prototypes, provisioning hardware and building out software architecture before deploying Pivotal's open-source Hadoop framework in a production environment. Today, Inovalon uses the technology to manage and distribute data; it also leverages the system's predictive analytics capabilities to improve patient outcomes.

Pivotal's software is delivering results through a partnership Inovalon has with pharmacy chain Walgreens. Walgreens' healthcare clinics across the country use software developed by Inovalon called ePass, which uses predictive analytics to identify conditions that patients might develop and flag gaps in care -- capabilities that Inovalon CTO Joe Rostock says wouldn't have been possible without Pivotal's business-grade technology.

"A database technology like Hadoop doesn't account for the distribution and management of data in the healthcare ecosystem," says Rostock. "However, Pivotal's suite of tools allows our data scientists to access and manage data, and our production engineers to run processing data through our analytics."

Security Concerns

From tools that simplify app development to layers of software that extract business insights, the Hadoop ecosystem is becoming much more business-friendly. Yet security concerns continue to hinder enterprise adoption of Hadoop. For starters, while traditional big data environments rely on perimeter security tools for protection, the very nature of Hadoop -- a distributed architecture -- adds a new layer of complexity, making it more difficult to secure.

What's more, data moves rapidly between Hadoop nodes, making it difficult for companies to establish role-based access controls and policies governing who can and can't tap into sensitive data. And finally, last-minute security add-ons to Hadoop have created a mind-boggling landscape for security-minded IT professionals.

"Apache Hadoop was initially developed without security in mind, and much of what we know as today's Hadoop security was added on just a few years ago," says Kevin T. Smith, technology director at Novetta Solutions and co-author of Professional Hadoop Solutions. "As a result, the security retrofit of Hadoop has resulted in a very complex approach that is confusing to many IT professionals who are not experts on security."

Fortunately, a growing number of vendors are stepping up with new and easy-to-use security tools. Consider, for example, Zettaset Orchestrator. A software application for big data management and security, Orchestrator now supports Hortonworks Data Platform 2.1. In addition to acting as a management layer that simplifies Hadoop deployments, Orchestrator delivers encryption, access control, policy enforcement and uninterrupted database uptime for greater security on Hadoop.

1 2 3 4 Page 2
Page 2 of 4
Bing’s AI chatbot came to work for me. I had to fire it.
Shop Tech Products at Amazon