Amazon Aurora is a managed, cloud-native relational database service with MySQL and PostgreSQL compatibility, designed to offer the performance and availability of commercial databases at a lower cost.
Aurora uses distributed architecture that separates compute capacity from storage, providing up to 128 TB of storage, high performance, and 99.99% availability.
Important concepts of Aurora provisioned and Aurora Serverless v2 databases include clusters and instances, advantages of this architecture, and capacity and pricing.
Aurora Serverless v2 scales the size of an Aurora instance based on workload and enables a richer feature compatibility with provisioned instances.
Capacity of Aurora Serverless v2 is defined by Aurora Capacity Unit (ACU), and pricing is based on ACU-hours, which ensures the most precise fit of resources to your workload.
Aurora Serverless v2 is available in all regions where Amazon Aurora is available, and the version support is similar to provisioned instances.
Aurora Serverless v2 ensures feature compatibility and upgrades are the same as for provisioned instances.
When planning for performance with Aurora Serverless v2, you will need to consider additional factors like minimum ACU, rate of instance scale, and buffer pool capacity, among others.
Amazon Aurora provides a cluster endpoint which always points to the current writer node in the cluster, and Aurora Global Database offers the ability to maintain a live copy of your database cluster in up to five additional AWS Regions.
Provisioned instances and Aurora Serverless v2 instances are complementary; choose which one is the best fit for your workload and validate these assumptions to optimize your database solutions on AWS.