Setup Overview
The core setup path takes a new environment from architecture planning through a locally verified ERDDAP, load-balancer integration, and final reboot testing. Complete the chapters in the order shown below.
The commands use placeholders such as YOUR_GCP_PROJECT_ID. Review and replace every placeholder before running a command. Never apply the example Terraform configuration unchanged or connect it to another group’s Terraform state.
Core deployment path
1. Project Architecture
Choose the serving model, identify required and optional components, assign ownership, and size the initial environment.
2. Cloud Infrastructure Setup
Provision the private VM, persistent disk, optional bucket, service accounts, IAM, networking, and snapshot policy with Terraform.
3. Accessing VMs
Connect through Identity-Aware Proxy without assigning a public IP or opening SSH to the internet.
4. ERDDAP VM Setup
Safely initialize storage, optionally mount GCS, install ERDDAP, validate it locally, and connect it to the load balancer.
5. Go-Live Checklist and Recovery Basics
Complete the minimum checks required before launch and understand when snapshots or restoration are needed later.
Optional processing infrastructure
ERDDAP does not require a processing VM or Cloud Run. Use these chapters only when the same group also develops or operates data-generation workflows.
Processing VM Setup
Optional cloud VM foundation for interactive development, reprocessing, backfills, and compute-heavy workflows.
Data Processing Templates
Two public, sanitized GitHub templates for Cloud Run Jobs and VM-based processing projects.
The later workflow and code sections are SWFSC implementation examples. They are useful references for processing design but are not part of the ERDDAP deployment sequence.