Scala and Spark for Big Data Analytics

Narrow Dependencies

When an RDD can be derived from another RDD using a simple one-to-one transformation such as a filter() function, map() function, flatMap() function, and so on, then the child RDD is said to depend on the parent RDD on a one-to-one basis. This dependency is known as narrow dependency as the data can be transformed on the same node as the one containing the original RDD/parent RDD partition without requiring any data transfer over the wire between other executors.

Narrow dependencies are in the same stage of the job execution.

The following diagram is an illustration of how a narrow dependency transforms one RDD to another RDD, applying one-to-one transformation on the RDD elements: