Running Kubernetes in a Manufacturing Line – What Could Possibly...
summary
Project implements Kubermatic to run Kubernetes-in-Kubernetes, provisioning a templated data‑center per site and deploying service clusters per site with Harbor, Argo CD, local S3-like storage and a containerized ISC DHCP to hand out IPs for printers and tablets. Architecture targets 99.96% SLA and multi‑site HA; current fleet is 39 clusters, 206 nodes, 878 CPUs, 3.2 TB RAM and 200 TB storage. Autoscaler tuning addresses Azure disk‑attach limits; design favors local-first failover and incremental rollouts—solution works in production after phased rollout.
AI-written summary · about
chapters (14)
- 0:00Intro
- 4:00Project Overview
- 10:30Team Structure
- 13:00Running Kubernetes in Manufacturing
- 13:10What can go wrong
- 13:57Wonderball
- 14:50Automation
- 16:55Kubernetes Clusters
- 22:05Failover Case
- 22:31Disaster Case
- 23:17What Went Wrong
- 30:13Platform
- 30:36Managed Service
- 31:18Outro
related talks
37mRunning Kubernetes in a Manufacturing Line
30mEdge Kubernetes and Healthcare: A Next-Gen Architecture that Scales Across Thousands of Locations
21mLessons from the Field: Running Kubernetes at the Wellsite
36mOptimizing Resource Usage in Kubernetes
01h 01mDeploying Kubernetes, a Container Cluster Manager
28mBuilding and Managing Kubernetes with Kubernetes
35mKubernetes journey
22mAnatomy of a production grade Kubernetes cluster