Multiple transformation and jobs setup

A transformation can have multiple data flows setup to together to run in parallel.

Due to memory restrictions and processing efficiencies, it is recommended that you limit the number of data flows within a transformation till 8 to 10.

A job can have multiple transformations or jobs that can run in sequence or in parallel. If any of the transformations fail, an email is send and the whole job aborts.

To ensure that each transformation is run in the job even when there are errors, re-evaluate the connection. Right-click the connection line, and evaluate the connection to Unconditional. This allows the next transformation to run in the line if there are error messages.

To set a job(s) and transformation(s) to run in parallel, right-click the step to run parallel steps after. Select the Run Next entries in parallel.

You can have conditional steps and unconditional steps (lock). For example to run table group 12 when group 6 fails and group 6 sends an error email that the job group 6 failed.