New Features in Revolution R Enterprise 6.2
Revolution Enterprise 6.2 leverages the latest stable release of open source R (2.15.3) and includes high-demand new capabilities companies require to extend predictive analytics across an organization while also optimizing the performance of sophisticated analytic models on big data sets. For example, customers with predictive models in production systems build thousands of models every week. At that scale of activity, traditional model-fitting techniques are simply too slow. Starting with a feature set of hundreds of candidate variables, Revolution R Enterprise users can now use feature selection techniques to reduce the number of variables, and then run stepwise regression to finalize the model for automated scoring.
Revolution R Enterprise 6.2 includes the following new capabilities:
High Speed Teradata Data Connection.Teradata is the first database for which Revolution R Enterprise has a dedicated parallel connection. Customers can seamlessly extract data from a Teradata database using the Teradata Parallel Transporter and write it to a high performance XDF format file, or simply analyze the data directly. The increased speed with which Revolution R Enterprise users can move the data saves a significant amount of time when working with a large dataset.
- Stepwise Regression for ‘Big Data’ Linear Models. This feature allows users to automate the process of building a model by using a rigorous method to test and select from among a range of variables that are available for use in the model. The result is a dramatic reduction in the total time needed to fit a model.
- Parallel Random Number Generation. This new functionality provides an R interface to the parallel random number generators supplied with the Intel MKL libraries. These allow high quality parallel random numbers to be used in distributed computations.
This update also includes the following improvements:
- A faster way to import or analyze fixed format text data
- Improvements to the linear, logistic and generalized linear modeling functions for enhanced control over interaction terms model fitting
- Sort, merge, and split enhancements that now by default make better use of available memory for improved performance
- Options to write by-group counts or summary statistics directly to a high performance XDF file for further analysis. Users have more control over the summary statistics that are reported