Aurora Pricing Calculator - Instances, Serverless & DSQL

Aurora is a cloud-native database with higher performance and availability. Aurora offers up to 5x throughput of MySQL and 3x of PostgreSQL on standard RDS. It automatically replicates data across 3 Availability Zones with 6 copies of your data. Aurora storage auto-scales up to 128 TB. While Aurora instance pricing is slightly higher than RDS, the improved performance and built-in high availability often provide better overall value.

Choose Serverless v2 for variable or unpredictable workloads. Serverless v2 automatically scales capacity based on demand, scaling in seconds and only charging for what you use. Instance-based is more cost-effective for predictable, steady-state workloads where you can right-size instances. Serverless is ideal for development environments, new applications, or workloads with significant idle periods.

Aurora DSQL is a serverless, distributed SQL database with PostgreSQL compatibility. It uses Distributed Processing Units (DPUs) as its billing metric, covering all database activity including compute, reads, and writes. DSQL scales to zero when idle with no hourly charges. The free tier includes 100,000 DPUs and 1 GB of storage per month. Multi-region clusters add approximately 50% to DPU costs for cross-region write replication.

Choose Aurora DSQL for distributed, multi-region workloads that need active-active availability with strong consistency. DSQL scales to zero and charges per DPU with no minimum. Choose Serverless v2 when you need MySQL compatibility, full PostgreSQL extension support, or when your workload is confined to a single region. DSQL is PostgreSQL-compatible but has a distinct feature set focused on distributed transactions.

I/O-Optimized is recommended for I/O-intensive workloads. Aurora Standard charges separately for storage ($0.10/GB) and I/O requests ($0.20 per million). I/O-Optimized has higher storage cost ($0.225/GB) but includes unlimited I/O. If your workload performs more than 1 million I/O requests per GB of storage per month, I/O-Optimized is typically more cost-effective and provides predictable pricing.

An ACU (Aurora Capacity Unit) represents approximately 2 GB of memory with corresponding CPU and networking. Serverless v2 scales between your configured minimum and maximum ACUs. Each ACU costs around $0.12-0.16/hour depending on engine and region. Set minimum ACUs to 0.5 for development or 2+ for production workloads that need faster scaling response.

Aurora Reserved Instances offer significant savings for long-term commitments. 1-year No Upfront terms save approximately 25-35% compared to On-Demand pricing. 3-year No Upfront terms save approximately 40-50%. Reserved Instances apply to instance-based Aurora only (not Serverless). They are tied to a specific instance type, region, and database engine family.

Yes, Graviton instances are recommended for most Aurora workloads. Graviton-based instances (db.r6g, db.r7g, db.t4g, db.x2g) offer up to 35% better price-performance than equivalent x86 instances. They're fully compatible with Aurora MySQL and PostgreSQL. Only use x86 instances if you have specific compatibility requirements.

Both engines offer similar Aurora benefits. Aurora MySQL is compatible with MySQL 5.7/8.0 and offers faster simple queries and replication. Aurora PostgreSQL is compatible with PostgreSQL and offers advanced SQL features, better JSON support, and extensions like PostGIS. Choose based on your existing expertise and application requirements.

Aurora provides free backup storage equal to your total Aurora cluster storage size (across all instances). Additional backup storage beyond this amount is charged per GB-month. Automated backups are retained for 1-35 days (configurable). Manual snapshots are retained until you delete them and count toward your backup storage usage.

Data transfer within the same Availability Zone is free. Cross-AZ replication traffic between Aurora replicas is free. Data transfer out to the internet is charged per GB after the first 100 GB/month free tier. Data transfer between Aurora and other AWS services in the same region is typically free. Plan for data transfer costs if your application frequently sends large result sets to clients outside AWS.

Aurora automatically replicates data across 3 Availability Zones. Your data is stored in 6 copies across 3 AZs, providing durability even if an entire AZ fails. Aurora can lose up to 2 copies without affecting write availability and up to 3 copies without affecting read availability. Failover to a read replica typically completes in under 30 seconds.