Problem
Paradox Shop's checkout fleet already publishes fraud-check jobs (arriving at 300/sec) to an isolated urgent pipeline and report jobs to a separate bulk pipeline; that isolation is correct. The bulk worker pool still has only 1 instance, well under the 1,200/sec arrival rate, so the 300,000-job backlog only grows instead of draining. Scale up the bulk pipeline, without touching the already-correct urgent one.
Required components
Click to add a component to the canvas.
Additional components
Browse by category. Click or drag a component onto the canvas.