AnsibleByExample

Red Hat AI llm-d — Enterprise GenAI Inference on OpenShift

Beginner
2024-01-012By Luca Berton

Lesson 1 of 15 in Foundations

Ansible Foundations to Production — full curriculum
  1. 1. Introduction to Ansible and Automation
  2. 2. Installing and Configuring Ansible
  3. 3. Running Ad-Hoc Commands
  4. 4. Writing Your First Playbook
  5. 5. Understanding and Using Modules
  6. 6. Variables and Facts
  7. 7. Conditionals and Loops
  8. 8. Handlers and Idempotency
  9. 9. Structuring Projects with Roles
  10. 10. Inventory Management
  11. 11. Using Ansible Vault for Secrets
  12. 12. Debugging and Troubleshooting
  13. 13. Real-World Automation Examples
  14. 14. Final Project: Deploy a Web Server
  15. 15. Bonus - What's Next with Ansible?
View path overview →

Introduction

Red Hat's llm-d transforms LLM inference into composable Kubernetes-native architecture with disaggregated serving and Model-as-a-Service tiers. This article covers the key announcements, architecture decisions, and practical implications for Ansible automation teams.

Key Highlights

What Changed

The Red Hat ecosystem continues evolving toward AI-driven, event-reactive automation. The 2026 updates focus on reducing manual intervention while maintaining governance and compliance.

Architecture Overview

# Modern Red Hat automation stack
platform:
  controller: AAP 2.7
  eda: Event-Driven Ansible
  hub: Private Automation Hub
  gateway: Unified Gateway (single entry point)

ai_layer:
  assistant: Automation Intelligent Assistant
  mcp: Model Context Protocol server
  rag: Bring-Your-Own-Knowledge

governance:
  approval: Human-in-the-loop gates
  policy: OPA/Gatekeeper integration
  audit: Full decision trail logging

Impact on Automation Teams

AreaBeforeAfter
Incident responseManual triageAI-assisted diagnosis
Playbook creationYAML from scratchNatural language + AI
EE buildingCLI onlyVisual self-service
Event handlingReactive scriptsGoverned pipelines
CompliancePeriodic auditsContinuous verification

Getting Started

# Check your current AAP version
ansible-playbook --version

# Install latest collections
ansible-galaxy collection install ansible.eda
ansible-galaxy collection install redhat.insights

Migration Considerations

  1. Version check — ensure AAP 2.5+ before upgrading to 2.7
  2. PostgreSQL upgrade — plan for PostgreSQL 13 → 17 migration
  3. Gateway architecture — consolidate entry points
  4. Collection updates — pin to compatible versions
  5. Training — upskill team on EDA and MCP concepts

Best Practices

  1. Start small — pilot AI features with non-critical workflows
  2. Keep humans in the loop — approval gates for production
  3. Invest in observability — you can't automate what you can't measure
  4. Document decisions — AI reasoning should be auditable
  5. Community engagement — participate in upstream development

Conclusion

Red Hat's llm-d transforms LLM inference into composable Kubernetes-native architecture with disaggregated serving and Model-as-a-Service tiers. These capabilities represent the next evolution of infrastructure automation — from imperative playbooks to intelligent, event-driven operations with AI assistance.

For a production-focused walkthrough, see Luca Berton's guide on Ansible Kubernetes day-2 operations.

All Tutorials