Skip to content

Latest commit

 

History

History
116 lines (77 loc) · 4.72 KB

README.md

File metadata and controls

116 lines (77 loc) · 4.72 KB
Raw

Deployment README.md

This folder contains the infrastructure-as-code and CI/CD pipeline configurations for deploying a conversational Generative AI application on Google Cloud.

The application leverages Terraform to define and provision the underlying infrastructure, while Cloud Build orchestrates the continuous integration and continuous deployment (CI/CD) pipeline.

Deployment Workflow

Deployment Workflow

Description:

  1. CI Pipeline (deployment/ci/pr_checks.yaml):

    • Triggered on pull request creation/update
    • Runs unit and integration tests

  2. CD Pipeline (deployment/cd/staging.yaml):

    • Triggered on merge to main branch
    • Builds and pushes application to Artifact Registry
    • Deploys to staging environment (Cloud Run)
    • Performs load testing

  3. Production Deployment (deployment/cd/deploy-to-prod.yaml):

    • Triggered after successful staging deployment
    • Requires manual approval
    • Deploys to production environment

Setup

Prerequisites:

  1. A set of Google Cloud projects:

    • Staging project
    • Production project
    • CI/CD project (can be the same as staging or production)

  2. Terraform installed on your local machine

  3. Enable required APIs in the CI/CD project. This will be required for the Terraform deployment:

    gcloud config set project $YOUR_CI_CD_PROJECT_ID
gcloud services enable serviceusage.googleapis.com cloudresourcemanager.googleapis.com cloudbuild.googleapis.com secretmanager.googleapis.com

Step-by-Step Guide

  1. Create a Git Repository using your favorite Git provider (GitHub, GitLab, Bitbucket, etc.)

  2. Connect Your Repository to Cloud Build For detailed instructions, visit: Cloud Build Repository Setup.

    Alt text

  3. Configure Terraform Variables

    Variable Description Required
    prod_project_id Production Google Cloud Project ID for resource deployment. Yes
    staging_project_id Staging Google Cloud Project ID for resource deployment. Yes
    cicd_runner_project_id Google Cloud Project ID where CI/CD pipelines will execute. Yes
    region Google Cloud region for resource deployment. Yes
    host_connection_name Name of the host connection you created in Cloud Build Yes
    repository_name Name of the repository you added to Cloud Build Yes

    Other optional variables include: telemetry and feedback BigQuery dataset IDs, log filters, sink names, service account names, bucket name suffixes, artifact registry repository name, and various role assignments for Cloud Run and CICD.

  4. Deploy Infrastructure with Terraform

    • Open a terminal and navigate to the Terraform directory:
    cd deployment/terraform
    • Initialize Terraform:
    terraform init
    • Apply the Terraform configuration:
    terraform apply --var-file vars/env.tfvars
    • Type 'yes' when prompted to confirm

After completing these steps, your infrastructure will be set up and ready for deployment!

Dev Deployment

For End-to-end testing of the application, including tracing and feedback sinking to BigQuery, without the need to trigger a CI/CD pipeline.

After you edited the relative env.tfvars file, follow the following instructions:

cd deployment/terraform/dev
terraform init
terraform apply --var-file vars/env.tfvars

Then deploy the application using the following command (from the root of the repository):

gcloud run deploy genai-app-sample --source . --project $YOUR_DEV_PROJECT_ID --service-account genai-app-sample-cr-sa@$YOUR_DEV_PROJECT_ID.iam.gserviceaccount.com

End-to-end Demo video

Watch the video