Bentley Automation Service jobs can be run in full, or incrementally.

Running a job in full means that all source documents called into the job's input set will be processed when the job is run. Running a job incrementally means that only the documents in the input set that require processing will be processed, and those that have already been processed will be skipped.

The first time you run any job, that job will run in full (whether you select to run it in full or incrementally), because none of the documents in the input set have been processed yet.

After running the job in full at least once, situations in which you want to run the job in full again include:

• if you have changed the parameters of the job definition
• if any documents in the job’s input set draw information from an external database

If a job has successfully completed and you run it again in full, the job starts over as if you had never run the job before, processing all the documents in the input set. Running a job in full also erases all history for that job.

A typical workflow might be to run a job in full at least once, and then run it incrementally as needed after that. You may decide to run a job incrementally:

• if documents in the input set have been modified since the job was last run
• if documents have been added to the input set since the job was last run
• if the job was stopped prematurely, before it had a chance to process all the documents in the input set