MongoDB also offers a tool to directly query Spark clusters and export data to MongoDB. Spark is a cluster computing framework that typically runs as a YARN module in Hadoop, but can also run independently on top of other filesystems.

MongoDB Spark Connector can read and write to MongoDB collections from Spark using Java, Scala, Python, and R. It can also use aggregation and run SQL queries on MongoDB data after creating a temporary view for the dataset backed by Spark.

Using Scala, we can also use Spark Streaming, the Spark framework for data streaming applications built on top of Apache Spark.