RDSAurora

Amazon Aurora, commonly referred to as Aurora, is a managed relational database engine within Amazon Web Services' RDS service. It provides MySQL-compatible and PostgreSQL-compatible databases with higher throughput, fault tolerance, and availability compared with standard open-source deployments, while remaining compatible with existing MySQL and PostgreSQL tools and applications.

Aurora uses a purpose-built distributed storage subsystem that is decoupled from compute resources. Data is automatically replicated across multiple availability zones, with six copies stored in three AZs. Storage scales automatically from gigabytes to as much as 128 terabytes, without manual provisioning. The architecture supports continuous backups to Amazon S3 and rapid failover to a healthy replica within the cluster.

Key features include automated backups, point-in-time recovery, and a global database option for cross-region replication. Read scaling is achieved with Aurora Replicas, and serverless configurations (Aurora Serverless) adjust capacity automatically for intermittent workloads. The compute layer scales independently of storage, supporting on-demand or provisioned instances.

Aurora is offered in two editions: MySQL-compatible and PostgreSQL-compatible. Applications can use standard drivers and SQL with minimal changes, though some features and versions differ from native databases. AWS maintains compatibility matrices and version support in its documentation.

Typical use cases include high-throughput transactional workloads, SaaS databases, and globally distributed applications needing low-latency reads. Considerations include cost, vendor lock-in, and potential feature gaps compared with on-premises databases, which can influence the choice between Aurora and other RDS engines.