RDDs support two types of operations:
Transformations
Actions
The transformation operation performs some functions and creates another dataset. Transformations are processed in the lazy mode and only those transformations that are needed in the end result are processed. If any transformation is found unnecessary, then Spark ignores it, and this improves the efficiency.
Transformations, which are available and mentioned in Spark Apache docs at https://spark.apache.org/docs/latest/programming-guide.html#transformations, are as follows: