Presently two ways are available to exploit the widely adopted Spark technology for carrying out ROOT based data analysis:

• PyROOTSpark : offers the possibility to use a simple map reduce interface to run existing root code with a few changes on Spark resources. The key is to use the PyROOT inteface in combination with PySpark.
• Spark-ROOT : offers the possibility to read ROOT files connecting the format directly to Spark's DataFrames/Datasets/RDDs