Route 53 Routing Policies: Weighted, Latency, Failover, and Geolocation (SAA-C03)
Route 53 has six routing policies and the SAA-C03 exam tests your ability to pick the right one for a given scenario. Each policy solves a different problem. This page explains when to use each and walks through a practice question that covers the most-tested distinction.
Short answer
The correct answer is B. Create latency-based routing records for each region and associate a health check with each record.
Latency-based routing directs users to the region that provides the lowest measured network latency from their location — this is determined by Route 53 using empirical latency data between AWS regions and user locations, not geographic distance. When you associate a health check with a latency record, Route 53 will only route to healthy endpoints, effectively implementing automatic failover as a byproduct. If one region becomes unhealthy, Route 53 routes to the next-lowest-latency healthy region. Option A (weighted) distributes traffic in a fixed ratio regardless of where the user is located — it does not optimize for latency. Option C (geolocation) routes based on the user's country or continent — a user in Germany always goes to eu-west-1 regardless of whether a different region is faster or closer at that moment. Option D (failover) only has a single primary and secondary — it doesn't scale to three active regions or optimize for latency.
The Question
A global e-commerce company has deployed application servers in us-east-1, eu-west-1, and ap-southeast-1. They want users to be automatically routed to the region that provides the lowest network latency from their location. If a region becomes unhealthy, traffic should automatically shift to the remaining healthy regions. Which Route 53 configuration achieves this?
Why B is correct
Latency-based routing directs users to the region that provides the lowest measured network latency from their location — this is determined by Route 53 using empirical latency data between AWS regions and user locations, not geographic distance. When you associate a health check with a latency record, Route 53 will only route to healthy endpoints, effectively implementing automatic failover as a byproduct. If one region becomes unhealthy, Route 53 routes to the next-lowest-latency healthy region. Option A (weighted) distributes traffic in a fixed ratio regardless of where the user is located — it does not optimize for latency. Option C (geolocation) routes based on the user's country or continent — a user in Germany always goes to eu-west-1 regardless of whether a different region is faster or closer at that moment. Option D (failover) only has a single primary and secondary — it doesn't scale to three active regions or optimize for latency.
Why the other options are wrong
Weighted routing splits traffic by a fixed percentage (e.g., 33%/33%/33%). It does not consider user location or network latency. A user in Tokyo would have equal probability of being sent to us-east-1 even if ap-southeast-1 is much closer.
Geolocation routing routes based on the DNS-resolved location of the user (country, continent, or US state). It does not measure actual network latency. A user in eastern Canada might be closer to eu-west-1 in terms of network latency, but geolocation would always send them to us-east-1.
Failover routing is an active-passive pattern with exactly one primary and one secondary. It does not support three simultaneously active regions or latency optimization. When the primary fails, all traffic shifts to the secondary regardless of user location.
Route 53 routing policies: the complete decision guide
Route 53 has six routing policies. Each maps to a specific use case. Simple routing: one record, one value (or multiple values returned randomly). No health checks. Used for single resources. Weighted routing: multiple records, each with a weight (0–255). Traffic is split proportionally. Weights of 10/10/10 give equal distribution; 10/0/10 sends nothing to the middle record. Used for: A/B testing, gradual traffic shifting between deployments, canary releases. Latency-based routing: Route 53 routes users to the AWS region with the lowest measured latency from their location. Multiple records point to different regions. Health checks integrate for automatic failover. Used for: global applications where performance matters. Failover routing: active-passive high availability. One primary record, one secondary. Route 53 routes to the primary when healthy, falls back to the secondary when the health check fails. Secondary can be another region, an S3 static site, or an EC2 instance. Used for: disaster recovery. Geolocation routing: routes based on user country or continent. Requires a default record for locations not covered. Used for: regulatory compliance (EU users must go to EU), content localization, language-based routing. Note: this is not latency — a user in Australia might get slower response if you force them to eu-west-1. Geoproximity routing (Traffic Flow only): similar to geolocation but can apply a bias to expand or shrink the routing radius of a region. Requires Route 53 Traffic Flow (paid feature). Used for: shifting traffic between regions with fine-grained geographic control. Health checks integrate with all policies except Simple. A health check monitors an endpoint (HTTP/HTTPS/TCP) or other Route 53 records. If a record is unhealthy, Route 53 routes to the next best option.
Ready to see how you'd score?
Take the free practice quiz and find out which AWS Solutions Architect Associate domains you need to focus on. No signup required.
Practice 5 similar questions
Same cert, same or adjacent domain. Use these after reviewing the explanation.