Amazon Aurora

Amazon Aurora is AWS's managed relational database, compatible with MySQL and PostgreSQL, built on a distributed storage layer that replicates across availability zones, with Serverless capacity scaling and global database options.

Replacing the storage engine rather than the database is the engineering idea that makes Aurora interesting. The SQL layer is MySQL or PostgreSQL, so applications and tools work unchanged, while underneath it a purpose-built distributed storage system handles replication and durability, which is where the performance and availability claims come from.

Compatibility is what makes that a low-risk choice, since a team can move an existing MySQL or PostgreSQL application without rewriting it and can, in principle, move back, which is a materially different proposition from adopting a proprietary database with its own dialect.

Serverless capacity is the option worth understanding before committing, because it scales the database with demand and suits variable or unpredictable workloads, while a steady production workload is usually cheaper on provisioned capacity. The Amazon global database option extends the same storage layer across regions for read scale and disaster recovery, and Performance Insights gives the query-level visibility a self-managed database would need a separate tool for. Aurora is metered by consumption with no simple rate table published.