Learning the basics about jobs

A Job is an entity made of Job entries linked by hops. These entries and hops build paths, telling the engine the sequence of inpidual tasks to accomplish. Therefore, we say that a Job is task oriented.

Graphically, Job entries are represented with small boxes, whereas hops are represented with directional arrows, as depicted in the following sample:

 Sample Job

The unit of execution inside a Job is a Job entry. Job Entries are grouped into categories according to their purpose, which can be transferring files, creating folders, sending emails, running scripts, and more. There is a main category named General, which contains many of the most used entries. In particular, this category includes the START Job entry, mandatory in every Job, as you can observe in the preceding image.

A hop inside a Job is a graphical representation that links two Job entries. The direction of the hop defines the order of execution of the linked entries. The execution of a Job entry does not begin until the one that precedes it has finished. This is distinctly different from Transformation executions.

While the execution of a Transformation is parallel and synchronous in nature, a Job executes in a sequential and asynchronous way.

A hop connects only two Job entries. However, a Job entry may be reached by more than one hop. Also, more than a hop may leave a Job entry.

A Job, just as a Transformation, is neither a program nor an executable file. It is simply plain XML. The Job contains metadata which tells the PDI engine which processes to run and the order of execution of those processes. Therefore, it is said that a Job is flow control oriented.