How open source Hadoop distributions foster innovation
Hadoop and big data capabilities are consistently touted as one of the most significant trends in business innovation, particularly as open source development allows users to implement new features. Some of the most used applications, such as Hadoop Hive, Pig and HBase, are the result of companies and developers contributing to the project.
According to a recent post by NoSQL blogger Alex Popescu, the open source architecture is central to Hadoop's success. Although the environment can be complicated, its open endedness allows developers and data professionals to exercise some latitude in driving innovations.
"[Hadoop] allows experimenting and trying out new ideas, while continuing to accumulate and storing your data," Popescu wrote. "It removes the pressure from the developers. That's agility. It's highly appreciated."
As a result of the innovation it enables, Hadoop has become the de facto tool for handling big data, analytics executive Matt Asay wrote in a recent ReadWrite column. Although legacy software vendors are increasingly trying to offer proprietary Hadoop products, this software does not hold the same promise for driving growth. Instead, organizations should continue to pursue open source deployments that enable them to work outside the confines of a vendor's walls and potentially unlock new, more creative applications for Hadoop and big data.