The Cluster Is Not the Platform: Operating Industrial Edge Kubernetes as a Fleet
summary
Claim: industrial edge Kubernetes becomes operable at scale by encoding per-site site profiles and exceptions, pushing them with GitOps (Argo CD) and local reconciliation, and managing upgrades in controlled lifecycle waves. Platform responsibilities include DNS and PKI (not IP-as-identity), local registries and replicated satellite services, preapproved network service patterns, and clear ownership boundaries. Success metrics are lead time, configuration drift (quality) and special-case rate, measured across upgrade waves.
AI-written summary · about
chapters (8)
- 0:00Introduction to industrial edge platforms
- 2:35Key platform components and challenges
- 5:26Differences between cloud regions and factories
- 8:18Standardizing edge sites with exceptions
- 11:50Operational workflow for fleet management
- 15:26Measuring fleet performance and outcomes
- 19:41Transitioning to a managed fleet
- 21:10Common anti-patterns in edge Kubernetes
related talks
06mEdge Kubernetes in Action: From Connected Vessels to Industrial Au...
28mOverview Of Challenges And Solutions For Orchestrating Applications To Multiple DC And Edge Clusters
10mDevelop once, deploy anywhere: flexible infrastructure extends to the edge
17mKeynote: Extending Kubernetes: Application Orchestration to 1000’s of Clusters in...
21mLessons from the Field: Running Kubernetes at the Wellsite
28mA Brief History of Kubernetes Fleet Controllers & Essential Features
07mKeynote: Pushing the Boundaries of Open Source for a Brand New Edge
09mSharpen Your Edge Three Ways With Kubernetes