Metrics and Worker Tuning
This chapter explains the infrastructure-facing operating pages in superadmin scope.
What belongs here
- host metrics
- Redis metrics
- Django or application metrics
- worker tuning and throughput review
Platform view matrix
| Area | What it helps answer | Typical next step |
|---|---|---|
| host metrics | is the host under unusual CPU or memory pressure | compare with queue or worker behavior |
| Redis metrics | is queue, cache, or memory pressure building up | decide whether backlog is forming |
| Django metrics | is application-side processing fresh or delayed | compare with worker cadence and recent changes |
| worker tuning | does throughput need controlled adjustment | tune only after reading the broader metrics picture |
Good operating pattern
- confirm whether the signal is local or platform-wide
- read host, Redis, and application metrics together
- treat worker tuning as a response to evidence, not as a first reflex