Error Budget Policy

Data artifactGovernance & ComplianceSafety, Security & Governancearc:ErrorBudgetPolicy

An organizational policy that binds error-budget state to engineering and release decisions: ship features while budget is healthy, prioritise reliability as it depletes, and freeze feature work once it is exhausted.

Responsibility. Specifies release and engineering-focus actions for each error-budget state.

Also known as: Error budget as organizational contract, SLO breach response policy

Direct neighbourhood (hover for relationship types)

Relationships

constrains control

Rollout Manager abstract Ch8.2A

Design guidance

MUST stop feature work when the error budget is exhausted (SLO breached) until reliability is restored.
SHOULD shift engineering focus to reliability work when the burn rate exceeds 2.0.
MAY ship features aggressively while the burn rate stays below 1.0.
SHOULD halt deployments and escalate to leadership on critical burn-rate alerts.

Quantitative guidance

As stated by the sources; verify before use.

Healthy: burn rate < 1.0; depleting: > 2.0; exhausted: SLO breach (99.9% SLO = 43.2 min/month budget) (Ch8.2A).

Classification

Patterns: Feature freeze on SLO breachReliability-velocity trade-off
Quality attributes: Reliability (ISO/IEC 25010 | NIST AI RMF: valid and reliable)Maintainability (ISO/IEC 25010)
Risks mitigated: Gradual reliability decay until a major outage forces emergency response

Sources

Ch8.2A: T. Nguyen, "Error Rates and Reliability," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 8.2A. ISBN: 9798244538229.