Autoscaling Controller  ·  View 30 of 37  ·  5 · Operations

Observability

Signal type by loop stage, and the alerts an operator can actually act on.

Editable source SVG draw.io All views
Signal plane Decision plane Actuation plane Managed capacity Metrics Ingest lag p99 target 5 s Decision latency p99 target 2 s Actuation latency p99 target 3 s Desired vs actual per target Health Signal freshness Lease held per shard Breaker state Time-to-serving distribution Quality Samples rejected stale or implausible Suppressed decisions by damping Partial actuations Warm-pool hit rate Alerts Stale past horizon Pinned at max 10 min 3 failed intervals Oscillation freeze Explain Signal history 13 months rolled Why this capacity journal only Actuation trace 1 : 1 to decision Cost of stability currency per day Observability — Signal Type by Loop Stage Every alert names a condition an operator can act on. "Why this capacity" is answered from the journal without replaying a single metric. v 1.0 · owner Data & AI Global Practice · date 2026-09

Decisions

  • Every alert names a condition with an action: pinned at max, actuation failing, signal stale, oscillation freeze, spend ceiling reached.
  • Suppressed decisions are a first-class metric. A controller that is doing nothing on purpose looks identical to one that is broken unless you count them.
  • "Why this capacity" is served from the journal without touching the metrics store.

Alert thresholds

  • Desired capacity pinned at max for over 10 minutes.
  • Actuation failing for 3 consecutive intervals.
  • A signal stale beyond its horizon; oscillation freeze; spend ceiling reached.

Per target

  • Current vs desired, time capacity-constrained, driving-signal saturation, decisions suppressed by damping, warm-pool hit rate, time-to-serving, cost.
◀  CI/CD and Policy Promotion All views Cost and the Price of Stability  ▶
Solution Architecture v1.0 · Data & AI Global Practice · 2026-09 · 37 views · Azure and open source30-observability