Interactive AIOps Demo

AIOps Incident Intelligence Lab

Synthetic telemetry and deterministic correlation rules only. No external API calls, no live Azure integration, and no real credentials.

Scenario and Time Window

Synthetic Telemetry

Synthetic logs, metrics, and alerts are generated locally from scenario templates and deterministic filters.

  • [METRIC] P95 Latency

    Azure Monitor | t-55m | 1200ms

  • [LOG] Dependency Slow Call

    Application Insights | t-52m | payments-api

  • [ALERT] Node CPU saturation

    Log Analytics | t-49m | 91%

  • [METRIC] Request Error Rate

    Azure Monitor | t-42m | 2.8%

Incident Timeline

  • Synthetic incident initialized for selected scenario.

Incident Lifecycle

new

Current assignee: Unassigned

Alert Correlation and Probable Cause

Correlation is driven by explicit deterministic rules. Any AI-style narrative below is mock explanatory text.

AKS latency cluster pressure correlation

High latency combined with node CPU saturation and dependency slow calls indicates capacity and downstream contention.

Evidence: e1, e2, e3

Contributing signals: P95 Latency, Node CPU saturation, Dependency Slow Call

Confidence limitations: No predictive model is used; this score reflects deterministic rule match strength. | Missing network trace depth can hide external dependency causes.

Alternative hypotheses: Ingress misconfiguration | Regional dependency degradation

Heuristic score note: Heuristic confidence score: 4/5 (rule-based, non-calibrated).

Mock AI-style narrative: Correlated telemetry suggests latency amplification due to compute pressure and slow downstream dependencies. (Marked as mock AI-generated text.)

Suggested Remediation and Approval

Simulated runbook actions require explicit human approval before execution.

  1. Confirm latency hotspot via Application Insights dependency traces.
  2. Scale impacted AKS workload and verify HPA thresholds.(Approval required)
  3. Review Azure Front Door / gateway routing before closing incident.

Approval status: Pending | Execution status: Not executed

Architecture Explorer

Demo: AIOps Incident Lab: Synthetic telemetry and deterministic incident correlation.

Implemented and verified

Data flow: Synthetic telemetry correlates into incident hypotheses, then approved remediation workflow advances lifecycle state.

Accessible alternative: use the component buttons below to inspect architecture details without zoom.

Legend

  • Ingress / Trigger
  • Orchestration
  • Data / Storage
  • Security
  • Observability
  • Operations
  • Synthetic Telemetry ~> Correlation Rules (scenario signals)
  • Correlation Rules -> Cause Panel (probable cause)
  • Correlation Rules -> Lifecycle Workflow (incident state)
  • Cause Panel -> Remediation Approval (recommended action)
  • Remediation Approval ~> Lifecycle Workflow (approval event)

Component Details

Synthetic Telemetry

Generate reproducible event streams for scenarios.

Responsibilities: Logs/metrics generation | Scenario shaping

Security controls: No real customer data | No live environment calls

Dependencies: Correlation Rules

Design decisions: Synthetic datasets keep the demo safe and deterministic.

Technical Details

Correlation logic is deterministic and testable. No trained predictive model is used.

  • Rule examples: latency + cpu + dependency slow-call, crashloop + restarts + exit codes.
  • Incident states: new -> acknowledged -> assigned -> investigating -> resolved -> closed.
  • Simulated remediation requires explicit approval before execution.

Sanitized Incident Payload

{
  "incidentId": "INC-AKS-HIGH-LATENCY",
  "scenario": "aks_high_latency",
  "status": "new",
  "syntheticTelemetry": true,
  "deterministicCorrelationRules": true,
  "selectedWindow": "30m",
  "correlatedRules": [
    "RULE-AKS-LATENCY-01"
  ],
  "remediationApproved": false,
  "remediationExecuted": false
}