• Logo
    Charmed Kubeflow
  • charmed-kubeflow.io/
  • More resources
    • Discourse
    • Mattermost
    • Matrix
    • GitHub
Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
Charmed Kubeflow documentation
Charmed Kubeflow documentation
  • Tutorials
    • Get started
    • Build your first ML model
  • How to
    • Install
      • General installation
      • Install in an air-gapped environment
      • Install on AKS
      • Install on EKS
      • Install on GKE
      • Install behind a web proxy
      • Install on NVIDIA DGX
      • Install using Terraform
    • Manage
      • Upgrade
        • Upgrade from 1.9 to 1.10
        • Upgrade from 1.8 to 1.9
      • Uninstall
      • Troubleshoot
      • Back up control plane
      • Restore control plane
      • Integrate with COS
      • Integrate with MinIO
      • Integrate with MLflow
      • Manage profiles
      • Configure High Availability for Istio Gateway
      • Configure Kubeflow Notebook creation page
      • Enable HTTPS
      • Enable Istio CNI plugin
    • Use
      • Configure advanced scheduling
      • Customise link configuration
      • Leverage PodDefaults
      • Use NVIDIA GPUs
      • Deploy NVIDIA NIMs
      • Launch NVIDIA NGC notebooks
      • Serve a model using Triton Inference Server
      • Deploy Autoscaling model serving
      • Perform inference on ISVCs using access tokens
    • Integrate with
      • Azure Blob Storage
      • Azure spot virtual machines
      • Identity providers
      • InAccel FPGA Operator
  • Reference
    • Release notes
      • Charmed Kubeflow 1.10
      • Charmed Kubeflow 1.9
      • Charmed Kubeflow 1.8
    • Supported versions
    • Monitoring
      • Prometheus metrics
      • Prometheus alerts
      • Grafana dashboards
      • Loki logs
  • Explanation
    • System architecture
    • Workload scheduling patterns
    • Security
      • Authentication
      • Authorisation
      • Cryptography
    • Charmed vs. upstream
    • MLOps tools
Back to top
Contribute to this page

ExplanationΒΆ

The following guides discuss concepts and features of Charmed Kubeflow (CKF).

Learn about CKF system architecture:

  • System architecture
  • Workload scheduling patterns

Read about supported security:

  • Security

Understand the differences between the upstream and CKF projects:

  • Charmed vs. upstream

Explore which MLOps are integrated within CKF:

  • MLOps tools
Copyright © 2025 CC-BY-SA, Canonical Ltd.
Last updated on Jul 16, 2025
Manage your tracker settings
Contents
  • Explanation