Modernizing Frontend Delivery: Migrating from ECS to AWS Amplify

Challenge

TargetBay’s BayRewards frontend is a React application that powers customer engagement features including coupons, progress tracking, and real-time leaderboards. Initially deployed on EC2 and later migrated to ECS, the application serves tens of thousands of global users. Despite the ECS migration, several operational and performance issues persisted:

• Regional Latency: Users outside the primary AWS region — particularly in APAC — experienced consistently slow load times, degrading the engagement experience that BayRewards is designed to deliver.
• Fragile Availability Model: A single ECS cluster served as the sole point of failure. Any cluster-level issue risked taking the entire frontend offline.
• High Operational Complexity: Managing Docker registries, task definitions, ECS service updates, and container health monitoring for a static React frontend added significant operational overhead disproportionate to the workload’s requirements.
• Persistent Compute Costs: The ECS deployment maintained fixed baseline compute costs around the clock, even during low-traffic periods when the application served minimal requests.

Solution

FactualMinds migrated BayRewards from ECS to AWS Amplify, replacing container-based hosting with a fully managed, edge-distributed static hosting platform purpose-built for frontend applications.

CI/CD and Deployment:

• Connected TargetBay’s GitHub monorepo to AWS Amplify Console
• Configured build settings (npm ci && npm run build) for consistent, reproducible builds
• Automated deployments triggered on push to main (production) and develop (staging) branches
• Enabled branch-based preview deployments for QA and design teams to review changes before merge

Hosting and Distribution:

• Amplify stores deployment artifacts in Amazon S3 with automatic versioning
• Amazon CloudFront (provisioned by Amplify) delivers the application from edge locations worldwide, eliminating regional latency
• SPA routing, redirects, and rewrites are handled natively by Amplify without custom server configuration

DNS and Security:

• Amazon Route 53 configured for DNS routing to bayrewards.io
• AWS Certificate Manager (ACM) TLS certificates provisioned and renewed automatically
• Optional AWS WAF integration available for additional request filtering

AWS Services Used

• AWS Amplify Console — Managed CI/CD, hosting, and preview deployments
• Amazon S3 (via Amplify) — Artifact storage with automatic versioning
• Amazon CloudFront — Global edge-cached content delivery
• AWS Certificate Manager — Automated TLS certificate provisioning and renewal
• Amazon Route 53 — DNS management and routing
• AWS WAF (optional) — Web application firewall for request filtering

Results

The migration from ECS to Amplify delivered improvements across performance, reliability, operations, and cost:

• Global edge delivery with CloudFront serving the application from edge locations worldwide, significantly reducing latency for APAC and other previously underserved regions
• Zero single points of failure as Amplify’s managed infrastructure eliminates cluster-level availability risks
• Simplified operations with no Docker images, task definitions, or ECS service configurations to manage — the entire deployment lifecycle is handled through Git push
• Reduced compute costs by eliminating persistent EC2/ECS compute in favor of pay-per-use static hosting with no baseline costs during low-traffic periods
• Automated CI/CD with branch previews enabling QA and design teams to review frontend changes in isolated environments before production deployment