Practices to Improve the Performance, Scale and Efficiency of Your Ansible Playbooks
summary
Ansible controller load comes from pushing modules and executing them on endpoints, so size controllers by CPU and memory (roughly 100 MB per fork and default forks ~5), enable memcache fact caching to avoid repeated facts gathering, use persistent connection plugins instead of short-lived local/paramiko, and partition work with instance groups and geographically isolated nodes so jobs run near devices. Use job slicing to break large inventories into parallel chunks; teams report scaling to ~15,000 network devices by combining these practices.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 2:24Problems of Scale
- 7:30Memory Requirements
- 11:08Other Considerations
- 13:31Common Strategies
- 15:52Scaling
- 16:40Plugins
- 19:05Vertical scaling
- 23:04Isolated nodes
- 25:13Defining isolated groups
- 25:44Job slicing
- 29:46Scaling issues
related talks
23mIntroduction to Ansible for Network Engineers
01h 34mAnsible 101
48mHow We Manage Infrastructure Across 1,000+ Servers Using ansible
37mAnsible 101
28mNo Compromise: Use Ansible properly or stick to your scripts
29mGood Practices and Security for your Ansible Automation Platform 2
21mA Distribution Framework over ANSIBLE
36mPulling the Puppet Strings with Ansible