Network Telemetry & AI-Driven Operations

High-Performance Networking & Cloud-Native · 30 illustrated lessons (text/HTML)

Outcomes

• Collect streaming telemetry and build a time-series pipeline
• Apply anomaly detection and root-cause analysis to network data
• Build a small telemetry collector and remediation workflow

Where we teach this

• OBS-301 · Telemetry and AIOps for Networks guided, self-paced

The lessons

Every lesson in this curriculum, in the order we teach it.

• 1Course introduction
• 2Telemetry fundamentals
• 3Modern network data collection
• 4Data pipelines and stream processing
• 5Choosing and using time-series databases
• 6Monitoring data visualisation
• 7AI-Ops fundamentals
• 8Anomaly-detection algorithms in practice
• 9Network traffic forecasting in practice
• 10Root-cause analysis (RCA)
• 11Intelligent alert consolidation
• 12Capacity forecasting and planning: ML-based resource forecasting and network expansion recommendations
• 13Automated fault remediation
• 14Telemetry data security
• 15Unified cloud-network monitoring: Telemetry in hybrid and multi-cloud, and a unified platform architecture
• 16Container and microservice monitoring: Kubernetes and Istio observability in practice
• 175G and edge monitoring: edge telemetry and a low-latency, high-reliability monitoring architecture
• 18Big-data platform integration
• 19Building real-time diagnostic tools
• 20Performance benchmarks and SLAs
• 21Case study: monitoring a large data-centre network in practice
• 22Case study: intelligent operations for a carrier's wide-area network
• 23Case study: low-latency trading networks in finance
• 24Model training and deployment
• 25High-availability system design
• 26Cost optimisation: data storage and compute resource management
• 27Open-source stack in practice: network monitoring and AI-Ops
• 28Comparing commercial telemetry solutions
• 29Hands-on: building a simple telemetry collector in Python
• 30Course summary and outlook