© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Most data scientists use nowadays functional or semi-functional languages like SQL, Scala or R to treat data, obtained directly from databases. Such process requires to fetch data, process it, then store again, and such process tends to be done outside the DB, in often complex data-flows. Recently, database service providers have decided to integrate “R-as-a-Service” in their DB solutions. The analytics engine is called directly from the SQL query tree, and results are returned as part of the same query. Here we show a first taste of such technology by testing the portability of our ALOJA-ML analytics framework, coded in R, to Microsoft SQL-Server 2016, one of the SQL+R solutions released recently. In this work we discuss some data-flow schemes for porting a local DB + analytics engine architecture towards Big Data, focusing specially on the new DB Integrated Analytics approach, and commenting the first experiences in usability and performance obtained from such new services and capabilities.

Peer Reviewed