Problem
Paradox Shop's checkout fleet serves 500 requests/second, publishing two kinds of background jobs: urgent fraud checks arriving at 300/sec that must start almost immediately, and bulk nightly report generation, which arrives at 1,200/sec with a 300,000-job backlog to clear. Sharing one queue and worker pool between them would let urgent jobs get stuck behind the bulk backlog. Build the fleet and a processing pipeline where urgent work is never delayed by bulk work.
Required components
Click to add a component to the canvas.
Additional components
Browse by category. Click or drag a component onto the canvas.