Why Virtualize Hadoop and How Project Serengeti Can Help
A very long post by Richard McDougall explaining why virtualizing Hadoop may make sense and how VMware’s Project Serengeti can help. Answering the question in the title, McDougall enumerates 6 reasons:
- Consolidation/sharing of a big-data platform
- Rapid provisioning
- Resource sharing
- High availability
- Security
- Versioned Hadoop environments
He’s also addressing two of the most common questions about Hadoop virtualization:
- Isn’t there a large performance overhead? Benchmark results are available in a whitepaper that can be read or downloaded below.
- Doesn’t vSphere use shared SAN storage only? (nb: the short answer is that vSphere supports both local and shared storage)
Original title and link: Why Virtualize Hadoop and How Project Serengeti Can Help (©myNoSQL)
via: http://cto.vmware.com/project-serengeti-theres-a-virtual-elephant-in-my-datacenter/
