The idea is:
Workflow-to-worker routing in Queue Mode
My use case:
We run n8n in queue mode with workers that have different hardware capacities, where some workflows are compute- and memory-intensive (e.g., large data processing, AI tasks, heavy transformations) and others are lightweight and latency-sensitive (e.g., webhooks, CRM updates, notifications). Currently, any worker can pick up any job, which means heavy workflows can occupy lower-capacity servers and cause resource contention, increased latency, and unpredictable performance for lighter automations.
We need the ability to route or tag workflows so that heavy executions are handled only by high-capacity workers, while lightweight workflows remain on the general worker pool, ensuring better performance, stability, and cost efficiency.
I think it would be beneficial to add this because:
I think it would be beneficial to add this because it would allow teams running heterogeneous worker infrastructure to better control resource allocation, prevent compute-heavy workflows from degrading overall system performance, and ensure latency-sensitive automations remain fast and reliable. By enabling workflow-level routing or worker tagging in queue mode, organizations could optimize hardware usage, improve stability under load, and scale more efficiently without needing to overprovision all workers to handle worst-case workloads.