Zonal redundancy replicates data across multiple availability zones within one region. It protects a Cosmos DB deployment when an entire zone becomes unavailable, keeping data accessible instead of placing all resilience eggs in one zone. section 11.1.2 section 11.4.2
Choose zonal redundancy by weighing local resilience against cost and regional coverage:
- Enable it with isZoneRedundant=True for each configured region when protection from zone-level failures matters. section 11.4.2
- Pair it with multi-region replication when the application also needs protection from regional failures and globally distributed access. section 11.1.2 section 11.4.2
- Accept higher storage and replication costs, and question the value in regions with fewer availability zones or existing regional redundancy. section 11.1.2
A resilient Cosmos DB configuration
A Cosmos DB account can assign Australia Central failover priority 0 and East US priority 1, then set isZoneRedundant=True in both regions. Each region distributes its replicas across availability zones, while failover priority selects the next region if the primary fails. section 11.4.2
C:/> az cosmosdb update --name alpacacosmosdbaccount --resource-group AlpacaRG --locations regionName=australiacentral failoverPriority=0 isZoneRedundant=True regionName=eastus failoverPriority=1 isZoneRedundant=True
Inspect the Azure CLI listing that enables zonal redundancy in both Australia Central and East US. section 11.4.2
