The Job object

The Job object specifies a map/reduce job's settings. Typically it is set in the run() method (Tool interface, described later). It should contain all necessary data to create a map/reduce SQL query. This data includes, but is not limited to:
  • database name
  • input table name
  • input column names
  • output table name
  • output column names
  • mapper/reducer/combiner classes
  • input/output key/value classes
  • partitioner class

Job is submitted using the JobRunner.runJob(Job) method. Before creating a SQL query and running it on the NPS, the framework performs job validation. If something is wrong with the job's configuration, an error message is logged.