I first learned about R last year at a local meetup on open science. http://www.meetup.com/Vancouver-R-Users-Group-data-analysis-statistics/events/73785912/

It made a lot of sense that R can act as a platform for sharing and validating science. Why not data science too?

Subsequently, I watched as R was adopted within various Big Data stacks (including Oracle, SAP):

Most recently, I saw a couple of enquiries about R and OLAP cubes:

http://stackoverflow.com/questions/16563782/r-clients-to-olap-mdx-servers

https://twitter.com/tanyacash21/status/355405308715220992

For a rather esoteric intersection of subjects, there seem to be interest in it.

And finally, I learned how R was used in a real-world problem at June’s Hadoop Summit’s training option on data science and machine learning.

So I thought to myself, what would it take to build something.

Welcome to X4R.

You can find it on github here.

https://github.com/overcoil/X4R

 

It’s a package for R to execute an MDX query (via XMLA) and to return the data-set into a data frame. You can use it to source data from cubes that you may already have. I’ve tested it with a few of our local sources include SAP BW, SAP HANA and SQL Server Analysis Services. Feel free to use it with any other XMLA sources that you are using. I’d love to have your feedback. One alternative that looks to be useful is to use OLE DB instead of XMLA so you can reach directly into SQL Server Analysis Services without any additional configuration.