Skip to main content

7. Deployment View

Infrastructure Overview​

CREstimate.ai is deployed on AWS using a containerized microservices architecture.

Deployment Components​

Container Orchestration​

  • Platform: AWS ECS Fargate (serverless containers)
  • Auto-scaling: Based on CPU/memory utilization
  • Load Balancing: Application Load Balancer with health checks
  • Networking: VPC with public and private subnets

Data Persistence​

  • Database: Amazon RDS for PostgreSQL with TimescaleDB extension
  • Cache: Amazon ElastiCache for Redis
  • Storage: S3 for static assets, logs, and backups
  • Backups: Automated daily snapshots, 7-year retention

Security​

  • Authentication: Auth0 managed service
  • Access Control: Cloudflare Access for investor/partner authentication
  • Encryption: TLS 1.3 in transit, AES-256 at rest
  • Secrets Management: AWS Secrets Manager
  • Network Security: Security groups, NACLs, WAF rules

Monitoring & Observability​

  • APM: Datadog for distributed tracing
  • Logs: CloudWatch Logs with structured JSON
  • Metrics: CloudWatch metrics + Datadog dashboards
  • Alerts: PagerDuty integration for on-call

CI/CD Pipeline​

  • Source Control: GitHub
  • CI/CD: GitHub Actions
  • Image Registry: Amazon ECR
  • Deployment: Blue-green deployments with automated rollback

Environments​

EnvironmentPurposeInfrastructure
DevelopmentLocal developmentDocker Compose
StagingPre-production testingScaled-down AWS (single instance)
ProductionLive systemMulti-AZ, auto-scaling

Detailed deployment diagrams and infrastructure-as-code coming soon...