Karthik Ranganathan - Hadoop Summit 2013 - theCUBE - #HadoopSummit
summary
Argues that a purpose-built converged appliance with local SSD CPU and RAM in the right ratio provides near-bare-metal Hadoop performance while removing hardware procurement and ops overhead, shipping a distributed file system and auto-rebalancing software that lets clusters scale out node by node and evolve as new workloads appear. Claims manageability gains by running Hadoop in VMs, offers Hortonworks-certified virtualization for Hadoop, and treats failure handling as a continuous engineering effort rather than a one-time fix.
AI-written summary · about
chapters (13)
- 0:00Introduction to theCUBE
- 4:22Guest praise for theCUBE
- 7:32Hadoop Summit live coverage
- 8:13Nutanix and converged infrastructure
- 12:24Mindset for enterprise scale
- 14:07Automation in enterprise vs hyperscale
- 15:08Hadoop and converged infrastructure
- 17:12Hadoop ecosystem growth and challenges
- 18:36Why join Nutanix from Facebook
- 23:07Platforma and modern BI
- 24:11Old vs new enterprise approaches
- 27:01Business value of new BI tools
- 28:49Changing business processes
