Spark R

In addition to native support for Apache Zeppelin, which is used primarily by Python and Scala developers, the R programming language can be used for statistical computing with InsightEdge. R provides a wide variety of statistical and graphical techniques to apply on the objects in the InsightEdge data grid.

This topic describes how to run R with the Apache Zeppelin web notebook.


R must be installed, along with any of its packages required for your purpose. To install R in Linux run the following command:

api-get install r-base

Running R from the Zeppelin Notebook

You can run R commands on the data grid from the InsightEdge Apache Zeppelin web notebook. Start the notebook and initialize the Spark context as described in Using the InsightEdge Web Notebook. You can view the R tutorial that is included in the notebook.



Before running the example below, install these 2 R packages with the following commands:

sudo R -e "install.packages('knitr', repos = '’)"
sudo R -e "install.packages('data.table', type = 'source',repos = 

Creating the Sample R Note

To read your data from the grid and write data back to the grid, create a new note with R commands similar to this sample:

//Reading salaries from the grid
df <- read.df("salaries", "org.apache.spark.sql.insightedge", header = "false", 
inferSchema = "true", na.strings = "NA") 
//Writing back as people to the grid
write.df(df, path = "people", source = "org.apache.spark.sql.insightedge", mode = "overwrite")