Another Possible Tool to use Hadoop with JMP: PivotalHD HAWQ?Team,
During a recent JMP visit with our friends at GE Power, we were asked to have a discussion on best ways to use Hadoop with JMP. We gave the previous advice of Stan and Dan to recommend first SAS Access/Integration for Hadoop as the optimal solution with the best speed and out of the box options, and then mention Hortonworks ODBC Driver for Hadoop as it has been tested successfully internally to some degree for a less optimal ODBC solution.
JMP and Hadoop
However it sounded like some teams within GE Power have found another possible tool that they claim works well with getting Hadoop data into JMP.
It is from PivotalHD and the product is called HAWQ. It is described as a Hadoop Native SQL Access Engine...see the attached the white paper and link below.
http://pivotal.io/big-data/pivotal-hdb
Sending this information in to Stan and Dan as we speak so they can check it out and give us some feedback on this. But in the meantime thought I would also share it with the field in case you also have heard of it from your customers and also so you can hear their responses on whether this is another viable or worthy tool option to be considered in the future for JMP and Hadoop.
Re: Another Possible Tool to use Hadoop with JMP: PivotalHD HAWQ?<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />Scott,
I took a quick look at the documentation.
At the core it is still using HDFS.
So we still could access either through the SAS Access engine or through an ODBC driver.
We have tested both both of the above options and both work without any issues when accessing HDFS data from Cloudera Horton Works or MapR or similar HDFS environments.
Best,
Stan
