How do I "do something" only at the end of all jobs when using Processing Concurrency?

I have a queue that is going to be handling **massive** amounts of data. I have set the processing concurrency to `2000`. What I want to do is iterate over all jobs in the _batch_, then when I have iterated over all I want to save to a database. Something like this pseudocode:
- Incoming batch of 2000 jobs
- Before iteration, initialize some arrays that will hold aggregated data
- Aggregate data from 2000 jobs into arrays
- Save aggregated arrays into DB

Any ideas?

The issue mentions Processing Concurrency, batches, and a queue backed by Redis, but names no files, tests, or entry points. Start by reviewing Kue's batch and concurrency behavior and the nine existing comments; done would require a clear, supported way to run aggregation after all jobs complete, or documentation explaining that behavior.

Written by the indexing model from the issue text.