Mike Olson, Cloudera - Hadoop World 2010 - theCUBE
Dave Vellante and John Furrier providing exclusive coverage at Hadoop World 2010, brought to us by Cloudera. The Hadoop movement is driving massive change, and leading the movement. They interview Mike Olsen, from Cloudera to get his opinion of where Hadoop and the future are going to bring us.
The men from The Cube ask Mike Olsen about the innovation, to which he speaks about the enterprise-readiness of the platform. "I think its interesting that Hadoop the software is an open source project that genuinely creates a new technology. It's not commoditized to the market, but its delivering new value, new power, doing new things for large scale enterprise data analysis," with the global community doing an incredible job of creating the framework and filling out the technology.
The talks that took place at Hadoop world highlighted the innovation and success of this new technology. 2010 was the second annual Hadoop world. John Furrier asked Mike Olsen what the biggest different is in 2010's Hadoop World from 2009's Hadoop World, and his response was the biggest difference has been the biggest single difference has been the arrival of the Hadoop platform. The core Hadoop project is at the center and has been complimented by new pieces that have been created by the community. This has allowed enterprises to roll it out into production eliminating the excess costs and time of having software employees, giving more availability of access to all types and sizes of enterprises. Together, the community has created the entire platform, incorporating data loading, interfaces, and integration.
Joe Vellante asks Mike Olsen about the survey data that he had spoken about while on stage, which was made of up 150 respondents. The data from the survey found that there are some 7000 nodes in the population, with the largest single cluster being 1300 nodes, "one enormous cluster, outside of the web companies," meaning NOT Google, Facebook or Yahoo! The data also showed that people think of Hadoop was a way to deal with a lot of data, with the power of the platform to deliver new analytical insight that before would have been impossible to have. Joe Vellante asks if he means Big Data now being accessible to a smaller company and Mike Olsen answers in the affirmative, saying this is what big data was called before there was a name for it, and "that at the end of the day you
have complex data from lots of different sources," and to make sense of it, you "need a new platform."