Amazon Aurora and Features
Learn about the fault tolerant architecture of Amazon Aurora and Aurora global databases.
We'll cover the following...
Before using Amazon Aurora in a production workload, it is important to understand that it is not a standalone engine. Aurora is compatible with MySQL and PostgreSQL, enhancing these engines’ availability, durability, throughput, and integration with AWS infrastructure.
Aurora databases can be deployed and managed through the Amazon RDS console, the AWS CLI, or AWS SDKs. However, unlike traditional Amazon RDS instances, Aurora organizes resources around a DB cluster, not a single instance. Aurora’s architecture ensures high availability through multi-AZ distributed storage even when a cluster contains only one DB instance.
Aurora cluster architecture
An Aurora cluster includes a shared, fault tolerant storage layer that spans multiple Availability Zones (AZs). This storage layer maintains six copies of data across three AZs, ensuring automatic redundancy and high durability, regardless of the number of DB instances running in the cluster. This architecture allows compute and storage to scale independently and ensures minimal impact on the application during failover or backup events.
By default, Aurora clusters include a primary DB instance for read and write traffic. To increase read throughput, additional read-only DB instances—called Aurora Replicas—can be deployed across AZs. Aurora supports up to 15 Aurora Replicas per cluster. All replicas share the same underlying storage, which reduces replication lag and eliminates additional storage costs.
In Single-AZ ...