Phases in Map Reduce

Map Reduce has 2 phases

• Input to each functions are key value pairs
• Map is a Mapper function and Reduce is the Reducer function

Mapper phase 

• First phase in the execution of map-reduce program.
• Data in each split is passed to a mapping function to produce output values.
• Several Map tasks are executed

Reduce phase

• Consumes output of mapping phase
• It consolidates the relevant record from match phase output.
• Several Reduce tasks are executed

Timeline of a map reduce job

Reduce phase may start before the end of map phase.