when the task at hand is too big for one person, then a crew or a team needs to be called in to complete the work.same is the concept of MapReduce,Typically a map reduction “team” would consist of one or more multi-processor computers(datanodes) and some type of master node(namenode) or program that manages the effort of dividing up the work between nodes (mapping) and the aggregation of the final results across all the worker nodes (reduction).
parallel processing is involved in MapReduce task.the data is processed in the form of (key,value) pair in mapreduce whereas,the policy of divide and conquer is applied. the above description is crux of understanding Mapreduce as there is more to learn about the mapreduce framework.