Intermediate

Cisco DNA Center AI Analytics

Explore how Cisco DNA Center uses AI-driven assurance to automatically detect network issues, perform root cause analysis, and guide remediation actions.

AI-Driven Assurance

DNA Center Assurance continuously collects and analyzes data from every device, client, and application on the network. Machine learning models process this telemetry to detect issues and provide actionable insights.

Assurance DomainWhat AI MonitorsKey Metrics
Client HealthOnboarding success, connectivity, experienceRSSI, SNR, throughput, latency, DHCP/AAA times
Network HealthDevice status, link utilization, errorsCPU, memory, interface errors, uptime
Application HealthApp performance, SLA complianceResponse time, packet loss, jitter, MOS score

Issue Detection and Resolution

  1. Anomaly Detection

    ML baselines learn normal behavior for each site, device type, and time period. Deviations trigger intelligent alerts with severity scoring.

  2. Issue Correlation

    Related alerts are grouped into issues. For example, multiple client failures in one area are correlated to a single AP or switch problem.

  3. Root Cause Identification

    AI traces the causal chain from symptoms to root cause, distinguishing between client-side, network, and application issues.

  4. Guided Remediation

    Step-by-step resolution actions are recommended with confidence scores, and many can be executed directly from the dashboard.

✅
Practical Tip: Enable the AI Network Analytics package in DNA Center to unlock advanced features like peer comparison, which benchmarks your network performance against anonymized data from similar deployments.

Advanced Analytics Features

AI Endpoint Analytics

Automatically profiles and classifies endpoints using ML, identifying device types, operating systems, and anomalous behavior without agents.

Rogue and aWIPS

AI-enhanced wireless intrusion prevention detects rogue access points, evil twins, and wireless attacks with reduced false positives.

Traffic Analytics

Encrypted traffic analytics uses ML to classify applications and detect threats within encrypted flows without decryption.

Trend Insights

Long-term trend analysis identifies slow-developing issues like gradual performance degradation or growing capacity constraints.

💡
Looking Ahead: In the next lesson, we will explore Cisco ThousandEyes and its AI-powered internet and cloud intelligence capabilities.