Does anyone have any experience/information on queue processor performance and/or any kind of metrics that  could be use to establish a baseline SLA for items processed per min/hour/day?

For background, we’re using an event based system for keeping case data synced with a backend SoR. As users complete workflows in Pega, the case is queued into a queue processor which will make a service call to the SoR and update any relevant response data in the case.

As we continue to build out the application, we’re getting questions about the timing of the background processing, and what kind of SLAs we can expect. Obviously performance is largely dependent on the actions being run by the QP, and performance testing will certainly give us more detailed answers there.

My goal here is to get a general idea of what's been seen previously before we build an entire application on top of queue processors only to find the throughput won't support our needs.