AWS Aurora Pricing: How To Optimize Costs In 2026

Amazon Aurora offers up to five times the throughput of standard MySQL and three times that of PostgreSQL. Its architecture combines the database engine with a cloud-native, SSD-based storage system built for high I/O operations to achieve this.

That said, AWS Aurora pricing can be a real headscratcher for customers, with concerns about the cost structure, pricing components, and ways to cut expenses.

In this post, we’ll demystify Aurora pricing and share tips to help you trim costs without sacrificing performance.

What Does Aurora Actually Cost?

Before diving into the details, here are ballpark monthly costs for common Aurora setups in us-east-1 to anchor your expectations:

These ranges vary heavily based on your I/O patterns, which is the variable most teams underestimate. Aurora is also not included in the AWS RDS Free Tier (which covers only standard RDS MySQL, MariaDB, PostgreSQL, and SQL Server Express).

What Is Amazon Aurora?

Amazon Aurora is an Amazon Web Services (AWS) relational database service. It offers higher performance and availability than standard Amazon RDS without significant changes to existing applications. It is also compatible with MySQL and PostgreSQL.

Aurora features a distributed, fault-tolerant storage system that automatically scales up to 128 TiB and supports up to 15 low-latency read replicas. This ensures robust data availability and recovery options.

Aurora’s architecture separates compute from storage at the infrastructure level. Your cluster has DB instances (compute heads that execute queries) and a shared cluster volume (where all data lives). Instances hold no persistent data themselves, and the storage layer replicates data across three Availability Zones using six copies automatically. Despite six copies existing, you pay for only one logical copy per region — a detail that matters when estimating your Aurora bill.

How Aurora Differs From Amazon RDS

We’ve shared an in-depth comparison of Amazon RDS vs. Amazon Aurora here. Below is a sneak peek at the key differences between the two AWS database solutions.

Amazon Aurora offers several advantages over standard RDS, including:

These added advantages come at a cost, making Aurora pricier than Amazon RDS pricing — at least at first glance. So, here’s how Aurora calculates your bill so you can identify what’s driving the charges and address them before overspending.

How Does AWS Aurora Pricing Work?

Aurora pricing is based on several cost factors, primarily your Aurora cluster configuration, database instance types, storage options, and I/O, as well as any optional features you decide to use.

The key pricing factors for Aurora are:

To keep this guide concise, here’s the TL;DR to help you pinpoint what you’re paying for and where to optimize Aurora costs. 

1. Pricing by Amazon Aurora cluster configuration

Aurora offers two configurations, and choosing between them is the single highest-impact cost decision you’ll make for an Aurora cluster:

The 25% rule: Check AWS Cost Explorer, filter by Amazon RDS, and group by Usage Type. Look for RDS:StorageIOUsage as a share of your total Aurora charges. If it sits above 25%, switching to I/O-Optimized can reduce your total Aurora cost by up to 40%. Below that threshold, Standard’s lower storage rate ($0.10 vs. $0.225/GB-month) makes it the better deal.

Here’s a worked example showing the impact. For a 1,000 GB database with heavy I/O (roughly 350 read pages/sec and 100 write pages/sec over 30 days):

Configuration	Storage Cost	I/O Cost	Total
Aurora Standard	~$129	~$233	~$362
Aurora I/O-Optimized	~$290	$0	~$290

In this scenario, I/O-Optimized delivers roughly 20% savings on storage and I/O costs alone. Switching between configurations is a cluster-level change that requires no data migration.

2. Aurora pricing by database instance types  

Charges are based on the type and number of database instances in your cluster.

One thing worth noting on instance families: Graviton-based instances (db.r7g, db.r6g, db.t4g) deliver better price-performance than their Intel equivalents at comparable or lower hourly rates. If you’re running Intel-based instances (like db.r6i) and haven’t evaluated Graviton, that’s a straightforward opportunity to reduce compute costs.

Note: Aurora typically costs more than standard Amazon RDS due to its superior performance.

3. Aurora pricing by storage and I/O operations

Aurora storage costs are based on the cluster configuration you choose (Standard or I/O Optimized) and the amount of storage you provision for your database (in gibibytes (GiB) per month).

For specific figures, the storage rate for Aurora Standard starts at $0.10 per GB-month, while I/O-Optimized storage runs $0.225 per GB-month. The I/O rate for Aurora Standard is $0.20 per million requests.

If you adjust your storage capacity during the month, your bill is prorated accordingly.

4. Backup storage

Backup storage pricing starts at $0.021 per GB-month. Aurora provides automated backup storage for free up to 100% of your provisioned cluster size — you’re only charged for backup storage that exceeds that threshold, plus any manual snapshots you’ve taken. Increasing your backup retention period or creating additional snapshots will raise your backup storage consumption.

5. Data transfer charges

Moving data from the internet into Amazon RDS is free. But data transfer to the internet starts at $0.09 per GB for the first 10 TB. Moving more than 150 TB per month drops the rate to $0.05 per GB. 

Consider this:

Image: Aurora data transfer fees

Here are a few other things to keep in mind:

6. Aurora Serverless v2 Pricing

Aurora Serverless v2 is an autoscaling instance type within a standard Aurora cluster. It scales by the second based on actual load, which makes it cost-effective for variable workloads and dev/test environments.

Serverless v2 bills on Aurora Capacity Units (ACUs), where one ACU corresponds to approximately 2 GB of memory along with corresponding CPU and networking resources. The cost for Aurora Serverless v2 is $0.12 per ACU-hour on Aurora Standard, or $0.156 per ACU-hour on Aurora I/O-Optimized.

For dev/test environments, setting the minimum ACU to 0 enables automatic pause. When the cluster has no active connections, compute charges drop to zero while storage costs continue, potentially reducing a development database to just a few dollars per month.

One important trade-off: Serverless v2 has no Reserved Instance equivalent. For sustained high-utilization workloads (running at near-peak capacity more than 60–70% of the time), provisioned instances with a 1-year Reserved Instance often work out cheaper. The hybrid pattern that works well in practice is a provisioned writer (stable baseline, RI-eligible) combined with Serverless v2 readers (flexible read scaling).

7. Additional Features

Activating optional features introduces extra costs. The features do add value, but you’ll want to evaluate whether you need them. Here are two examples:

8. Additional Features

Activating optional features introduces extra costs. The features do add value, but you’ll want to evaluate whether you need them. Here are three examples:

Now that we’ve covered AWS Aurora pricing, let’s talk about optimizing your Aurora costs.

Hidden Costs That Surprise Aurora Teams

Beyond the pricing components above, a few charges consistently catch teams off guard:

I/O bills that exceed compute costs. For OLTP workloads with frequent small transactions, I/O is often the largest single line item on the bill. At $0.20 per million I/Os, a busy production database generating 1 billion I/Os per month accumulates $200 in I/O charges alone, on top of compute. Check your I/O percentage in Cost Explorer (the 25% rule above) and switch to I/O-Optimized if warranted.

Orphaned snapshots from deleted clusters. When you delete an Aurora cluster, manual snapshots are not automatically deleted. They continue billing at standard backup storage rates indefinitely. This is particularly common in dev/test environments where clusters spin up and down frequently. Audit with aws rds describe-db-cluster-snapshots --snapshot-type manual and look for snapshots whose cluster identifiers no longer match active clusters.

RDS Extended Support fees. Running an end-of-life Aurora engine version (such as Aurora PostgreSQL 12, which reached community EOL in February 2025) triggers Extended Support charges of $0.10 per vCPU-hour in years 1–2, escalating to $0.20 per vCPU-hour in year 3. On a db.r6g.2xlarge (8 vCPUs), that adds roughly $584/month in Extended Support fees alone, on top of normal instance costs. Plan major version upgrades before EOL dates.

Tips For Estimating And Optimizing Your Aurora Costs

To estimate, control, and optimize Amazon Aurora costs, consider these immediately actionable strategies:

1. Right-size your DB instances

Choose the appropriate instance size based on your workload requirements. Start with smaller instances and monitor performance using tools like Amazon CloudWatch to avoid overprovisioning, which can increase unnecessary costs. AWS Compute Optimizer can analyze your CloudWatch metrics and flag over-provisioned, under-provisioned, or idle instances with specific recommendations.

2. Use RIs where applicable

For predictable workloads, consider Reserved Instances (RIs). Committing to a one or three-year term can save you up to 66% off the On-Demand rate. RIs are size-flexible within the same instance family and region.

If your infrastructure is likely to change (different instance families, regions, or engines), Database Savings Plans offer up to 35% savings with a 1-year commitment. They apply automatically across Aurora, RDS, DynamoDB, ElastiCache, and other database services, giving you more flexibility than RIs at a slightly lower discount.

3. Take advantage of Aurora Auto Scaling

Activate Aurora Auto Scaling to adjust your read replica capacity based on actual demand. This feature ensures you only pay for needed compute, preventing over-provisioning charges during low-traffic periods.

4. Monitor your I/O operations

Track your I/O usage since Aurora charges for read and write operations on Standard clusters. Use Amazon CloudWatch metrics (VolumeReadIOPs and VolumeWriteIOPs) to analyze I/O patterns. Apply the 25% rule: if I/O exceeds 25% of your total Aurora spend, switch to I/O-Optimized to eliminate those charges entirely.

5. Optimize backup storage costs

Regularly review your backup retention policies. Aurora automatically backs up data, but unnecessary manual snapshots (especially from deleted clusters) can quietly accumulate storage costs. Define a clear data retention policy that aligns with your business needs, and automate snapshot cleanup as part of any cluster deletion workflow.

6. Use data compression

This will reduce your storage requirements, significantly lowering your storage costs while maintaining performance.

7. Conduct regular cost reviews

Schedule a routine to review your Amazon Aurora costs. If you run a small operation, you can use AWS Cost Explorer and AWS Budgets. Analyze spending patterns, identify cost spikes, and adjust your strategies for ongoing cost optimization.

But when you need deeper insights, like Cost per Customer, Cost per Request, or Cost per Feature, to uncover who, why, and what’s driving your AWS spend (including Aurora costs), pick a powerful cloud cost optimization platform like CloudZero.

With CloudZero, you’ll get:

CloudZero’s unique approach has earned the trust of brands like Skyscanner, Shutterstock, and Remitly. In fact, Skyscanner saved enough in three months to cover their annual subscription. It’s your turn to take control. to see CloudZero in action.

Pricing information last verified May 2026. Features, pricing, and availability may have changed. Please verify current details with AWS before making decisions.