Microsoft R Open / Revolution R Enterprise Reviews & Insights

We implemented Microsoft R along side Windows Subsystem for Linux with Airflow to do ETL data loading last year through until late this year. We had some success with both products but due to limitations to Airflow and R not being completely fully functioning versus open source R suites we have cancelled both tools not long ago. I can say during its time it was easy to set up, R was easier to use to send processing jobs through than other solutions before and after. It just did not meet our needs once we got ETL loading to a more robust point.

If you are a MS shop specifically, or have more generic data requirement needs from Microsoft sourced data this will work well. If you have a lot of disparate data across a number of unique platforms/cloud systems/3rd party hosted data warehouses then this product will have issues or a lack of documentation on the net. Performance-wise this product is equal to other R platforms out there.

My organisation has 40+ services ranging from 00s to millions of users, therefore iterations, changes and forecasting must be done in a controlled, rigorous and statistically appropriate manor as well as using tools that are efficient and provide analytics the ability to interrogate data en masse. R provides this.

The main use of R for us is in structural forecasting and joining data together.

Using R to forecast gives our analysts the flexibility to interrogate a lot of data to give better modelling accuracy. This includes plugging in per second and minute data over multiple years for services with 100s of millions of interactions per year. Our seasonality means we can often shoot up from 00's to millions of users in a matter of hours, so knowing this will help allocate resource, both server, helplines and support.

Microsoft R is used by specific users when parallel processing is desired. Otherwise, most are fine with Base R.

Well suited if parallel processing is desired, eg. giant datasets or complex calculations. Otherwise, R studio might be sufficient. In either case, some level of programming in R will be required.

Microsoft R is typically used for statistics and data analysis needs for marketing data and developing machine learning algorithms. It helps solve optimization and developing predictive models. R is currently being used by one person in the department considering the size of the company, but looking to add more users of R sooner or later.

It's well suited for developing machine learning algorithms and statistical/data analysis. However, analysis in R does take some time so if you're churning a high number of projects, you might want to consider something else.

I have used Revolution Analytics Rev-R enterprise 7.0 for data analytics project. I was also engaged in beta-testing release D as well. Rev-R actually solves a Big Data gap by allowing data scientists to load big data in Hadoop HDFS and run complex algorithms such as Random Forest or decision trees by running the algorithms in a distributed way on the cluster. That helps to draw insights from big data sets without having to script complex programs in say Java or Python.

Revolution Analytics is a very compelling product for Big Data Analytics. It allows distributed computing over multiple hadoop nodes thus allowing HDFS to do its role cleanly i.e. cheap massive storage and it does good job of running algorithms using R or similar programming language on Hadoop. It would be definitely advantage for the organization who uses either R or SAS as their statistical model development tool as Rev-R support both the platforms. Overall, very positive experience with Rev-R.