Anomaly-driven Operational Dictionary
An operational dictionary stores thresholds and norms that anomaly detection and stewards use to interpret signals.
Why this is needed
- Without norms, every spike looks like an incident—or nothing does.
- Controlling and ops need shared thresholds.
- Fits EntiHub anomaly scheduling and steward triage.
Proposed entities in EntiHub
Governance settings
| Setting | Recommended value | Why |
|---|---|---|
approvalRequired |
true |
Threshold changes affect alerting behavior. |
historyEnabled |
true |
Compare norms across periods. |
allowSameAuthorApproval |
false |
Maker/checker separation for sensitive changes. |
deletedRetentionDays |
90 (or higher) |
Soft-delete retention for auditability. |
Entity definition (YAML sketch)
name: ops_threshold
displayName: Operational threshold
domain: Operations
description: Operational threshold
version: "1.0.0"
approvalRequired: true
historyEnabled: true
deletedRetentionDays: 90
displayRefColumn: threshold_code
uniqueConstraints:
- name: uq_ops_threshold_threshold_code
columns: [threshold_code]
columns:
- name: threshold_code
dataType: string
nullable: false
maxLength: 64
- name: metric_name
dataType: string
nullable: false
maxLength: 128
- name: entity_name
dataType: string
nullable: true
maxLength: 64
- name: min_value
dataType: decimal
nullable: true
- name: max_value
dataType: decimal
nullable: true
- name: severity
dataType: int
nullable: false
- name: owner_team
dataType: string
nullable: false
maxLength: 128
- name: status
dataType: string
nullable: false
maxLength: 16
YAML sketches
Sketches highlight the important fields and governance flags. Trim or extend columns to match your domain. See Setup Entity for the full YAML reference.
Sample data
| threshold_code | metric_name | entity_name | min_value | max_value | severity | owner_team | status |
|---|---|---|---|---|---|---|---|
| TH-CREDIT-CHG | credit_limit_weekly_changes | customer | 50 | 2 | Credit | active | |
| TH-IBAN-NULL | iban_null_rate_pct | supplier | 2 | 3 | Treasury | active |
Benefits for the company
- Early warning before reports or pipelines break.
- Shared language between ops and stewards.
- Feeds anomaly configuration intentionally.