CODEWITHSUNDEEP
hadoop
Mital Awachat
Mital Awachat

Reputation: 87

MapReduce execution sequence on Single Node

I am learning Hadoop.
I am running Hadoop on single node.
According to my knowledge Reducer runs after completion of Mapper (and it makes sense as well).
But when i ran MapReduce job on 200MB file, Reducer started before completion of Mapper. I didn't use any Combiner.
Can anyone explain why?

enter image description here

enter image description here

Upvotes: 0

Views: 146

Answers (1)

Thomas Jungblut
Thomas Jungblut

Reputation: 20969

The reduce phase involves copying and merging the output of the data from the mappers to the reducer.

Since copying and merging intermediate outputs does not need a barrier (you don't need to wait on all mappers to complete), that's what the reducer is doing while the mappers run.

Upvotes: 1

Related Questions