SYSTEM Cited by 3 sources

Terraform

Terraform (HashiCorp) is an infrastructure-as-code tool: declare cloud resources in a domain-specific language (HCL), terraform apply reconciles the actual state to declared state. Terraform is a common platform-layer config primitive across AWS, GCP, Azure, and on-prem.

Stub page — expand on future Terraform-internals sources.

The self-hosted, enterprise distribution is Terraform Enterprise (TFE) — see that page for its AWS reference deployment and the customer-operated multi-Region DR pattern.

Seen in

• sources/2026-09-18-aws-how-csiro-built-scalable-cost-optimized-genomic-variant-querying-on-aws — Terraform defines the entire sBeacon deployment, shipped as a container that sets up the dev environment. The per-institution self-contained deploy (terraform apply / terraform destroy) is what makes sBeacon's decentralization model work: each institution runs the full stack in its own AWS account with no shared infrastructure and no cross-account trust. terraform destroy tears down the service (created datasets lost, but not the underlying VCFs). (Source: sources/2026-09-18-aws-how-csiro-built-scalable-cost-optimized-genomic-variant-querying-on-aws)

• sources/2026-09-09-aws-validating-multi-region-dr-for-terraform-enterprise-with-aws-fis — Terraform's S3-stored state files are the substrate of the circular-dependency lesson: failover scripts that read terraform output from primary-Region state break during a regional S3 outage. See systems/terraform-enterprise and concepts/circular-dependency.

Seen in (Terraform-on-ECS)

• sources/2024-08-08-figma-migrated-onto-k8s-in-less-than-12-months — Figma used Terraform on ECS to define service task-set templates, which paired with a separate deploy step that cloned the template and substituted image hashes. The two-step pattern was the source of many "forgot the order" bugs and motivated the move to a single-source service definition on EKS. Post-migration fast-follow: moving more resources out of Terraform into AWS Controllers for Kubernetes (ACK) for further stack unification.