Skip to main content
In this lesson we cover Amazon Route 53 Application Recovery Controller (ARC) — a managed service that continuously validates standby readiness and provides programmatic routing controls to orchestrate DNS-based failover across Regions and Availability Zones. ARC reduces the manual steps and uncertainty involved in multi-region failover by combining health/readiness checks with API-backed routing controls that integrate with automation tools. Overview
  • ARC continuously verifies that backup or standby locations are operational, scaled, and configured to accept traffic when failover is required.
  • Define readiness thresholds (for example, if application health degrades beyond a defined percentage). To automate failover when thresholds are breached, integrate ARC routing controls with automation such as Amazon CloudWatch alarms, Amazon EventBridge, AWS Lambda, or Systems Manager Automation to invoke routing control changes. Manual diversion is also supported for planned maintenance.
  • Amazon Route 53 performs DNS routing. ARC routing controls act as the programmatic switch to update how Route 53 routes traffic between active and standby endpoints.
How ARC works (high-level)
  1. Define Cells, Recovery Groups, and Resource Sets that represent your application topology and cross-region resources.
  2. Configure Readiness Checks that continuously validate resources in cells and resource sets.
  3. Evaluate readiness results and, when thresholds are breached, invoke ARC Routing Controls (manually or via automation).
  4. Routing Controls update Route 53 configuration (via your automation) to divert traffic to the standby site.
  5. ARC logs and records changes for auditing and operational visibility.
Active / Standby example
  • Active site: US‑East‑1 hosts the primary application and serves production traffic.
  • Standby site: US‑West‑1 hosts a scaled standby deployment able to accept traffic during failover.
  • ARC runs readiness checks on the standby environment to ensure it is ready to receive traffic.
  • When failover is required, ARC routing controls are used together with Route 53 and your automation to update DNS and point traffic to the standby load balancer.
Architecture example
  • US‑East‑1 (active): Application Load Balancer (ALB) + EC2 instances serving traffic; application reads/writes to a global DynamoDB table.
  • Standby region: Equivalent ALB + EC2 instances connected to the same global DynamoDB table.
  • Route 53 routes traffic to the active ALB, and ARC routing controls plus automation update DNS to route to the standby ALB during failover.
A technical diagram titled "Application Recovery Controller Architecture" showing an active US‑EAST‑1 region and a standby US‑WEST‑1 region inside a recovery group. It highlights routing control, readiness checks, resource sets spanning cells, and AWS icons (Route 53, database) to illustrate failover and readiness monitoring.
Key concepts Use cases
Use ARC to gain continuous visibility into the recoverability of your application stack and to automate safe, auditable routing changes during failover or maintenance windows.
Operational considerations
  • Automation: Integrate ARC with CloudWatch Alarms, EventBridge rules, Lambda functions, or Systems Manager Automation runbooks to create repeatable failover flows.
  • Testing: Regularly run simulated failovers to validate readiness checks, resource scaling, DNS propagation, and any automation runbooks.
  • Auditing & safety: Use ARC’s control-plane audit logs and implement multi-actor approval in automation (where required) to prevent accidental routing changes.
  • Latency & data consistency: Consider application-level data replication (e.g., multi-region DynamoDB) and eventual consistency models when failing over between Regions.
Summary Amazon Route 53 Application Recovery Controller simplifies multi-region failover by continuously validating standby environments and exposing API-backed routing controls to update DNS routing safely. Remember the core building blocks: cells (independent failover units), recovery groups (collections of cells), resource sets (cross-cell resources), readiness checks (operational validation), and routing controls (programmatic DNS/traffic switching). Links and references

Watch Video