Revolution Analytics Upgrades Revolution R Enterprise Analytics Software

Bookmark and Share

Revolution Analytics, a provider of software, services and support for the open source R project, has released upgrades to its commercial-grade analytics software built upon the powerful open source R statistics language for R-based enterprise-class data analytics. Revolution R Enterprise 6.2, available now, features new advances in high-performance predictive analytics, including quick and easy stepwise regression for big data linear models and predictive analytics on data stored in Teradata databases.

Revolution R Enterprise 6.2 features new capabilities that extend predictive analytics across organizations while optimizing the performance of analytic models on big data sets. “We specialize in making R run faster, be more scalable especially with big data, and to be enterprise-ready so it can integrate into IT systems for production applications and provide all the support, training and services companies need to integrate R into their operations,” David Smith, Revolution Analytics’ VP of marketing and community, tells 5 Minute Briefing. With a feature set of hundresds of candidate variables, Revolution R Enterprise users can use feature selection techniques to reduce the number of variables and run stepwise regression to finalize the model for automated scoring.

New in version 6.2 is a high speed Teradata data connection, the first database for which Revolution R Enterprise has a dedicated parallel connection. “Somebody who’s an analyst working on a server running Revolution Enterprise can ingest data from the Teradata system and then build predictive models on it with Revolution Enterprise. We’re using Teradata’s proprietary Teradata Parallel Transporter. With this high-speed connection you can ingest data six times faster than before,” Smith explains. The greater speed with which customers can move data saves a significant amount of time when working with large datasets.

 Revolution R Enterprise 6.2 also features stepwise regression for big data linear models, allowing users to automate model building by using a method to tests and select from among a range of variables that are available for use in the model. Parallel random number generation provides an R interface to the parallel random number generators supplied with the Intel MKL libraries, providing high quality parallel random numbers for use in distributed computations. Updated RevoDeployR web deployment framework features new APIs for script management and new priority scheduling features that improve the management and operation of deployed R routines. Updated Java, JavaScript and .NET client libraries provide support for application developers, making it easier to integrate on-demand R-based computations with desktop, web-based and mobile apps.

Version 6.2 of Revolution R Enterprise is available now. For more information, visit