What happened to HadoopLink?

Questions
, , , , ,
1

These days engineers can’t escape the notion that Industrial Internet of Things (IIoT) and Big Data are going to revolutionize automation, so I thought it would be a good idea to learn about Hadoop using WL. In 2013 Wolfram announced ‘Mathematica gets big data with HadoopLink’, so I thought it would be instructive to play around in a Notebook.

To my disappointment it seems Needs[HadoopLink`] died on the vine. Can someone from Wolfram explain why?

If Wolfram are serious about WL being a credible platform for Data Analytics it will need to provide us with a way of accessing and computing with Big Data. Perhaps the strategy is to get Data Analytics working with Little Data first? And perhaps after that a free developers kernel can be run on the slave nodes? Whatever the game plan, it would be good to know where we’re heading.

2

Here is a response from Tech Support. My bad!
Has anyone been using it ?

Hello Dr. Robert Prince-Wright,
Thank you for contacting Wolfram Technical Support regarding issues loading HadoopLink.
Have you installed HadoopLink from the GitHub repository as well as its dependencies? According to the blog post (Mathematica Gets Big Data with HadoopLink—Wolfram Blog), you will need the following to run HadoopLink:

  1. HadoopLink folder from: GitHub - shadanan/HadoopLink: Mathematica package for reading files off of HDFS · GitHub
  2. Apache Ant
  3. Mathematica (version 7 or higher)
  4. Wolfram Workbench (for building the documentation notebooks)
  5. Hadoop version 0.20, patched to include the typed bytes binary data support for Hadoop Streaming.

Please let me know if you come across any issues installing HadoopLink and those dependencies.