Using Terraform Enterprise to Support 3,000 Users
summary
Booking.com describes running HashiCorp Terraform Enterprise with 3,000 active users by auto-creating GitLab repos, TFE workspaces and OpenStack projects from a central service that generates Terraform code. The pipeline uses Packer images, custom Terraform providers, Sentinel checks and Puppet, and supports mass updates of modules to all workspaces. Provisioning drops from ~45 minutes to under five minutes with many VMs ready in about 30 seconds, and a factory model bootstraps 100 apps in three months.
AI-written summary · about
chapters (13)
- 0:00Introduction and context
- 1:46Problem and high-level solution
- 3:30User scale and legacy issues
- 6:20Ownership and automation goals
- 7:43System architecture overview
- 9:07Demo of VM provisioning
- 12:20Advanced workflows and CLI use
- 14:47Group machine management
- 16:31Workspace creation process
- 19:19Workspace management demo
- 21:44Centralized updates and scaling
- 23:31Scaling to 3,000 users
- 24:12Results and achievements
related talks
29mInfinity and beyond: Scaling Terraform for the global enterprise
19mSynchronization of Infrastructure Provisioning & Application Deployments
55mBest practices
21mAccelerating cloud transformation with Terraform: A World Bank success story
29mEverything, Everywhere, All at Once: Terraform and Indeed's Move to AWS
27mAdvanced Multipurpose Infrastructure: A Self-Managed Cloud Environment with HashiCorp Terraform
25mScaling Terraform at
29mEphemeral Cloud Infrastructure for Acceptance Tests with Terraform Enterprise & GitLab
