1

When I look over the documentation for Route 53 Routing Policies they list 7 different policies. How should I pick the right one if I am building out new infrastructure in AWS for a US-based e-commerce website? Here is the list of offerings:

  • Simple
  • Failover
  • Geolocation
  • Geoproximity
  • Latency-based
  • Multivalue Answer
  • Weighted

These different routing policies are pretty confusing. I would of course always want failover for my DNS. Why would I ever choose Simple? Is there some major difference between Geolocation and Geoproximity? What do they mean by "Traffic Flow Only" next to Geoproximity?

Because of the confusion around all of this I am for now choosing latency-based routing. That seems simple and intuitive. I have ELBs to handle failover for me, and with my latency routing I have associated health-checks, so it seems like my bases are covered.

Notes:

  • We have a handful of ELBs but they are mostly in us-west-2 region, some in us-east-1.
  • These ELBs are serving the same site out of multiple regions.
1

To answer your question, ill first try to give simple explanations for each policy to help with the confusion (https://docs.aws.amazon.com/Route53/latest/DeveloperGuide/routing-policy.html).

  • Simple: a basic routing policy that randomly sends a user to one of your addresses.

  • Failover: You have a primary (active) route and a secondary (passive) route. When the primary fails, it routes to the passive. If passive fails, you have an issue.

  • Multivalue: Multiple routes that each have health checks. It will randomly select an address (Simple routing policy) from the healthy routes. A more complex failover policy.

  • Geolocation: Route users dependent on where they are located. If a user is in England and gets better latency in Germany, they will still be routed to England.

  • GeoProximity: A more fine grained Geolocation. You would be able to route people located on the border of France and Germany (based on longitude) to a specific route. "Traffic Flow Only" is to specify that you must use the Route 53 traffic flow to configure this policy. It is the most complicated of the route 53 policies

  • Weighted: This policy allows percentage based balancing. Routing 30% of people to route-1, 20% to route-2 and 50% to route-3.

  • Latency: Route to the address that gives the user the best latency.

Now for your architecture, your current implementation would handle instance failures with the ELBs and the latency based routing (with health checks) would handle the routing failures. If you don't require routing your users to the lowest latency address, use the multivalue policy instead.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.