Multi-Region Active-Passive diagram template

Builds a new scene in your browser. Your existing scenes are not touched.

About this design

Active-passive keeps a second region ready to take over without paying for two fully live stacks. The primary region serves all traffic; its database replicates continuously, and object storage is copied across regions, so the standby always holds recent data. The standby region runs a scaled-down copy of the application, enough to start quickly but small enough to be affordable. DNS with health checks watches the primary and, when it fails, shifts traffic to the secondary, which is first promoted and scaled up. Two numbers define the design: recovery point objective, how much data you can afford to lose, set by replication lag, and recovery time objective, how long failover takes. Use the template to discuss drills, because an untested failover is a hope, and the human decision about when to fail back.

Diagram as text

This is the source of the diagram, in the ArchBoard diagram DSL. Paste it into Tools, Diagram from text to rebuild or change it.

title "Active-passive multi-region"
direction LR
user "Users" -> dns route53 "DNS with health checks"
dns-with-health-checks -[active]-> lb elb "Primary LB" -> service primary "App (region A)" -> db rds "Primary DB"
dns-with-health-checks -[standby]-> lb elb "Standby LB" -> service standby "App (region B)" -> db rds "Replica DB"
primary-db -[async replication]-> replica-db