Comment by hobos_delight

…yes - processing 3.2G of data will be quicker on a single machine. This is not the scale of Hadoop or any other distributed compute platform.

The reason we use these is for when we have a data set _larger_ than what can be done on a single machine.