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
Origins In Automated Network Discovery
- IP Fabric began as consultants automating network discovery to avoid re-documenting messy customer networks.
- They turned that automation into a product to provide accurate, repeatable network maps.
Automation Enables, Twins Enable Autonomy
- Automation is necessary but limited to scoped closed loops; autonomy needs global understanding.
- Digital twins can provide the broader view required to move from automation to autonomy.
Start With Forwarding-Plane Modeling
- A network digital twin models the forwarding plane so you can test behavior without touching production.
- Simulating full control planes is currently impractical, so modeling forwarding/state is the pragmatic starting point.