The Everything Feed - All Packet Pushers Pods TNO042: Building a Network Digital Twin for Automation and AI (Sponsored)
Sep 19, 2025
Darren Fullwell, Field CTO of IP Fabric, shares his extensive networking experience, from early LANs to modern enterprise architectures. He delves into the fascinating concept of network digital twins, explaining their importance in automation and how they simulate network behavior to predict changes effectively. Fullwell discusses the role of AI in enhancing network operations and emphasizes the need for observability and adaptability in NetOps careers. His insights highlight the transformative potential of these technologies in networking.
AI Snips
Chapters
Transcript
Episode notes
Collect The Right Layers For Path Validation
- Capture physical, L2, and L3 topology plus control-plane data relevant to forwarding.
- Use that combined model to validate paths and test security and policy impacts before changes.
Combine Twin History With Telemetry
- Time-series history in the digital twin helps pin root causes by comparing past working states to current failures.
- Pair the twin's structural baseline with near-real-time telemetry from observability tools.
Guide Observability With The Twin
- Use the twin's topology to narrow the telemetry search space and guide AIOps correlation.
- Focus data collection on elements that affect packet paths rather than dumping all vendor-specific details into big storage.