Cross Cluster Execution of Argo Workflows
summary
Demonstrates running the Argo control plane in one cluster while actual workflow pods execute in separate execution-plane clusters using kubeconfig and a service account for RBAC, with Metaflow compiling Python flows into Argo workflows and Argo Events wiring triggers. Shows a working demo: a 3‑step hello flow and a first.py → second.py trigger‑on‑finish chain that runs across clusters. Claims this saves cost and improves isolation but requires automation to create kubeconfigs and manage credentials.
AI-written summary · about
chapters (8)
- 0:00Cross-cluster Argo workflows intro
- 1:02Metaflow and Argo integration
- 4:33Cross-cluster workflows explained
- 6:22Control vs execution planes
- 11:57Demo: cross-cluster execution
- 14:42Argo Events for event-driven
- 16:45Demo: event-triggered workflows
- 18:58Q&A
related talks
20mDemystifying Argo Workflows: An Architectural Deep Dive
25mOne Engine To Rule Them All: Unifying Cloud Workloads With Argo Workflows
52mCloud-Native Kubernetes Workflows on AKS with Argo
52mCloud-Native Kubernetes Workflows on AKS with Argo
36mAutomating Cloud-native Spark Jobs with Argo Workflows
18mFrom Click To Chaos: Linking Argo Workflows and Backstage for Automate...
27mBuilding a Data Science Platform with Argo Workflows
25mHow to Multiply the Power of Argo Projects By Using Them Together