Determining number of reduces

Question

I'm running a MapReduce job on a hadoop cluster but I'm confused in what determines the number of Reduces that run a given MapReduce job on a cluster running MapReduce v1 (MRv1)?

Can someone help?

Answer 1

Number of reduces are set by the developer itself.

The right number of reduces seems to be 0.95 or 1.75 * (nodes * mapred.tasktracker.tasks.maximum).

At 0.95 all of the reduces can launch immediately and start transfering map outputs as the maps finish. At 1.75 the faster nodes will finish their first round of reduces and launch a second round of reduces doing a much better job of load balancing.

Hope this helps.