DevOps in Infrastructure — When It's Mostly Ops...
summary
A DevOps infrastructure case study about inheriting three datacenters and 4,000 physical servers managed with Terraform and containers, facing long hardware lead times, poor cabling, and a prolonged software upgrade effort. The team reorganized from separate build/run silos into product-aligned service teams, introduced visible service catalogs and tentative pricing, and adopted a hybrid strategy: retain a small (20%) metal portfolio for sensitive telecom workloads while steering developers to orchestration and public cloud for cost-effective, mature applications.
AI-written summary · about
chapters (10)
- 0:00Presentation by Renaud
- 0:45Presentation of the team
- 3:37Why infrastructure?
- 10:22What we led
- 12:25Light on every floor
- 13:47Dev eats the world
- 16:53How do we face the giant?
- 20:42How do we give infrastructure the letters of nobility?
- 25:25Cultural change
- 29:14End
related talks
30mDevOps: Supporting aircraft from hanger to take-off
36mDevOpsSup - Extended DevOps approach
36mDevOps in Outsource… Really!
27mOur DevOps Journey So Far (and Terraforming)
27mCase Study: Re-Thinking Our Infrastructure Tooling
49mExponential Organizational Growth with Infrastructure as Code
18mScaling DevOps in Complex Environments: Challenges, Failures, and Successes
28mDevOps in a Data Warehouse: Inside Out