MapR Review

It differentiates itself from other Hadoop distributions because of the lack of namenode, the base components are implemented in C, and the NFS functionality.


Valuable Features

MapR is one of three leading Hadoop distributions. The key differentiators I found the most valuable are:

  • Lack of namenode (no SPOF)
  • Performance – the base components are implemented in C
  • NFS functionality – copying to MapR FS is as easy as copying files to a directory

Improvements to My Organization

I haven't yet used it in production environments, but I've used it extensively in two proofs-of-concept.

Room for Improvement

It would be nice to have new developments in the Apache space (Spark, Storm, etc.).

Deployment Issues

I've had no issues with deployment so far.

Stability Issues

It's been stable so far.

Scalability Issues

There have been no scalability issues so far.

Customer Service and Technical Support

Customer Service:

I haven't yet had to use customer service.

Technical Support:

I haven't yet had to use technical service.

Previous Solutions

I tested both Cloudera and Hortonworks, and I rate it similarly to Hortonworks and slightly higher than Cloudera.

Initial Setup

Initial setup is rather straightforward thanks to detailed documentation covering all the bases. More advanced installations required manual deployment to an extent, but this too is well-documented.

Implementation Team

I implemented it myself.

ROI

I haven't put it into production yet.

Other Advice

I don't have any advice because the installation documentation is superb.

Disclosure: My company has a business relationship with this vendor other than being a customer: We have partnerships with all leading Hadoop vendors.
Add a Comment
Guest

Sign Up with Email