7.83 Latency Routing Strategy

The latency-based routing strategy redirects each user to the AWS Region that delivers the lowest network latency from their location. Imagine an instance on the US West Coast and another on the Australian East Coast: a user in South America is routed to the US one because it is the closer of the two, while a user in South-East Asia is sent to the Australian one. Latency between users and a Region is computed by Route 53 from real measurements made within that Region.

If no Region offers a closer alternative, Route 53 still picks the best of the available options. Users in Europe and Africa, with only US and Australian endpoints, will be served by the US instance because it yields the lower latency. This strategy is the right choice when latency is the primary user-experience driver.

Set it up in Route 53

  • Create a record with the Latency routing policy.
  • Name it (e.g. latency.mon-domaine.com), type A, default TTL.
  • Endpoint 1: Ireland instance IP, Region eu-west-1, record ID latency-Ireland.
  • Endpoint 2: Frankfurt instance IP, Region eu-central-1, record ID latency-Germany.
  • Endpoint 3: Tokyo instance IP, Region ap-northeast-1, record ID latency-Japan.

Optionally attach a health check to each endpoint (covered in the next module). Once the records are saved, hitting the URL from, say, Paris routes you to Frankfurt because that Region offers the lowest latency from Paris — connecting through a Tokyo VPN would route you instead to the Asian endpoint. This is the strategy to pick whenever response time matters most to your users.

Summary

Latency-based routing strategy in AWS Route 53 automatically directs users to the server instance with the lowest latency, typically the geographically closest one. This routing method is configured by creating multiple records pointing to instances deployed in different regions (e.g., USA, Europe, Asia), with Route 53 evaluating latency metrics to deliver optimal user experience. Users are intelligently routed based on real measured latency from their region, ensuring faster response times for latency-sensitive applications.

Key points

  • Latency-based routing directs users to instances with the least latency, typically those closest geographically
  • Route 53 evaluates latency per region and routes traffic accordingly, improving user experience for latency-sensitive applications
  • Configuration involves creating multiple records for instances in different regions (e.g., Ireland EU-West-1, Germany EU-Central-1, Asia), each tagged with its respective region
  • Users from different parts of the world are automatically routed to the nearest optimal instance—European users to European instances, Asian users to Asian instances, etc.
  • The strategy leverages AWS's ability to measure actual latency metrics from users in each region to make intelligent routing decisions

FAQ

What is latency-based routing strategy in AWS Route 53?

It is a DNS routing policy that automatically directs users to the server instance with the lowest latency. Route 53 measures latency from different regions and routes traffic to instances that provide the best performance for each user's location.

How do I configure latency-based routing in Route 53?

Create multiple records with a latency-based routing policy, each pointing to an instance in a different region. For each record, specify the instance's IP address and its AWS region. Route 53 will automatically evaluate latency and route accordingly.

When should I use latency-based routing?

Use latency-based routing for applications where user experience is sensitive to latency—such as interactive applications, real-time services, or where response time significantly impacts user satisfaction. It works best with instances deployed across multiple geographical regions.