Merv Adrian - BigDataNYC 2013 - theCUBE - #BigDataNYC
summary
Claim: 7% of enterprises have moved past pilots to deployed Big Data projects. Hadoop shifts from why and where to how and who, with distributions in the hype-cycle trough while SQL-on-Hadoop starts delivering interactive queries. Practical blockers are governance and security; installers use Kerberos and Accumulo (and projects like Knox), pick secondary stores (HBase, Oracle connectors, MongoDB, Basho) for frequent use, and favor deferred schema plus hybrid cloud as the operating model evolves.
AI-written summary · about
chapters (10)
- 0:00Introduction to BigDataNYC
- 0:32Hadoop ecosystem growth in 2013
- 2:56Hype cycle and trough of disillusionment
- 5:03Security challenges in Hadoop
- 8:11SQL on Hadoop debate
- 10:17Choosing data stores for use cases
- 12:04Data platform and distributed mainframe
- 14:55Hybrid cloud and software-defined data centers
- 16:20Balancing opportunities and enterprise readiness
- 17:43Closing remarks and bumper sticker summary
