Core Concepts and Overview of Aurora Serverless V2
Aurora Serverless V2 introduces a highly elastic and cost-effective way to run Aurora databases, abstracting away server management.
Aurora Serverless is an on-demand, autoscaling configuration of Amazon Aurora. It automatically starts, scales, and shuts down database capacity based on application needs, eliminating the need for manual database instance management and capacity planning, especially for unpredictable workloads. With Aurora Serverless, you don’t manage database instances; instead, you define a capacity range and AWS handles the rest.
The fundamental unit of capacity for Aurora Serverless V2.
Technical Specs: ACU Composition (V2): Blend of CPU, memory, and networking. ACU Size: Approximately 2 GB of RAM per ACU.
Aurora Serverless V2 automatically adjusts database capacity to match application demand.
Technical Specs: Scaling Range (V2): Configurable from 0 ACU to 256 ACUs. Scaling Granularity (V2): Increments as small as 0.5 ACUs.
Aurora Serverless V2 can scale down to 0 ACUs during inactivity, which pauses the database and stops compute billing. The database resumes automatically upon a new connection. However, when it goes into hibernation, it needs to come out of hibernation, which is not an instantaneous process; it takes a while to come back out of that, unlike Lambda where things happen and shut down instantaneously.
Billing for Aurora Serverless V2 is based on actual usage.
Technical Specs: Charged per second of use. You pay on a per-second basis for the database capacity that you use. You still pay for the storage when it's not in use; when it goes into hibernation, although you're not paying for database capacity, you are still paying for the storage that is there ready for it when it comes out of hibernation.
Aurora Serverless V2 Architecture and Features
Aurora Serverless V2 is designed with specific architectural components and features to provide high performance, scalability, and integration with the AWS ecosystem.
Aurora Serverless V2 leverages a warm pool of ACUs managed by AWS, allocating capacity based on workload and configured minimum/maximum ACU settings.
Mixed Clusters
Supports creating a single Aurora cluster with a mix of provisioned and serverless v2 instances.
Example
Provisioned writer for steady production, serverless v2 read replicas for spiky read traffic.
Data API
Allows database interaction over HTTPS, making it suitable for serverless applications.
Functionality
Handles connection pooling.
High Availability
Aurora Serverless V2 is designed for high availability.
Deployment
Supports Multi-AZ deployments.
Disaster Recovery
Aurora Serverless V2 integrates with Aurora Global Database for multi-region disaster recovery.
Failover
Serverless v2 replicas can automatically adjust their capacity during a failover event.
Behavior
Serverless v2 replicas can automatically scale up to match writer capacity during failover.
Aurora Serverless V2 Use Cases
Aurora Serverless V2 is particularly well-suited for workloads with unpredictable demand and for environments focused on cost efficiency and rapid development.
Ideal for applications that have infrequent, intermittent, or unpredictable traffic patterns.
Offers significant cost reduction by pausing instances when not in use, making it very good for developers who want to quickly deploy something and have it shut itself down when not in use.
Provides a cost-effective solution for multi-tenant architectures where per-tenant databases need to scale independently.
Comparison: Aurora Serverless V2 vs. Provisioned Aurora Cluster
comparison-table
Understanding the differences between Aurora Serverless V2 and traditional Provisioned Aurora clusters is crucial for selecting the appropriate deployment model for varying workload characteristics.
This table highlights the key differences in capacity management, maximum capacity, and ideal workload types.
| Option |
Capacity Management |
Maximum Capacity |
Suitable Workloads |
| Serverless V2 |
Automatic scaling of compute in 0.5 ACU increments based on traffic. |
Up to 256 ACUs (approx. 512 GB memory). |
Infrequent, intermittent, or unpredictable workloads. |
| Provisioned |
Manual selection and management of fixed instance classes (e.g., db.r6g.xlarge) based on vCPU and RAM. |
Wider range of high-end instance sizes for demanding workloads. |
Steady, predictable workloads. |
Aurora Serverless V2 Integrations
Aurora Serverless V2 can be integrated with other AWS services to enhance its capabilities, particularly for global deployments.
Aurora Serverless V2 clusters can be part of an Amazon Aurora Global Database, providing a single Aurora database that spans multiple AWS regions. This setup is designed for global applications requiring low-latency local reads and robust disaster recovery capabilities. It supports both Aurora Serverless v2 and Provisioned clusters.