Running PIG Scripts - Hadoop
Running PIG Scripts - Hadoop #OPENBOXEducation
There are two modes for running the PIG Scripts
Local Mode : in this mode, MapReduce is locally simulated with the local JobRunner class of Hadoop.
In this mode all the Pig Scripts will be executed on a single machine without having any need for Hadoop MapReduce and HDFS.
This mode is considered when there are smaller sets of data for testing the code. Since , it is most appropriate and fastest mode.
The PIG program executes in context of a local JVM and access the data through the local existing file system of the machine.
This mode helps in executing the pig programs against subsets of data.
MapReduce Mode (or) Hadoop Mode: in this mode the pig programs will run on an Hadoop cluster. Before running them on the cluster the programs will be transformed into MapReduce jobs.
This mode helps in executing the pig programs against data sets unlike local mode.