The circle of life. We used to track jobs in a db, but that would eventually hit performance issues (locks, noisy neighbors, etc). So more nosql solutions and s

The circle of life. We used to track jobs in a db, but that would eventually hit performance issues (locks, noisy neighbors, etc). So more nosql solutions and services showed up. New devs gobbled up the concepts and APIs but were frustrated that distributed nosql solutions were SaaSified and wanted to bring control back in house, in a single db, and we are back to square one. But with better tooling and hardware in theory. Hopefully you have a dedicated db at least to avoid noisy neighbor workloads but you may still have task queue work that hits scaling limits on the single db causing tasks to interfere with one another--so you can shard the db and isolate task streams, and the circle continues