Pricing overview

Amazon Aurora DSQL is the fastest serverless distributed SQL database with virtually unlimited scale, the highest availability, and zero infrastructure management. It automatically scales compute and storage to match your workload demands, so you only pay for what you use with no upfront costs. Your monthly bill is based on two primary components: database activity, measured in Distributed Processing Units (DPUs) and storage, measured in GB-month. With the AWS Free Tier, you can get started on Aurora DSQL for free.

DPU pricing

Aurora DSQL charges for all database activity using a billing unit called the Distributed Processing Unit (DPU). You can think of a DPU as a measure of how much work the system does to run your SQL workload. This includes compute resources used to execute query logic (e.g., joins, functions, aggregations) as well as the input/output (I/O) required to read from and write to storage. Aurora DSQL tracks this work in real time and bills only for active usage. When your cluster is idle, usage automatically scales to zero and you incur no DPU charges.

If you configure a multi-Region cluster, Aurora DSQL automatically replicates writes to each peered cluster in other AWS Regions. These replicated writes incur extra DPU charges equal to the cost of the originating writes.

To help you monitor and understand your DPU usage, Aurora DSQL breaks down and reports DPU activity to Amazon CloudWatch as ComputeDPU, ReadDPU, WriteDPU, and MultiRegionWriteDPU. The sum of these DPU sub-components approximates the total DPU usage shown on your monthly bill.

Storage pricing

Storage is billed based on the total size of your database, measured in GB-month. Aurora DSQL ensures your data is highly available and strongly consistent by replicating it across three Availability Zones within an AWS Region. You are only charged for one logical copy of your data per Region.

If you configure multi-Region clusters, storage is billed in each Region where your data is replicated, meaning you will see charges for every peered cluster in another AWS Region.

Free tier

Aurora DSQL includes a free tier to help you get started with no upfront cost. Each month, your first 100,000 DPUs and 1 GB of storage are free and automatically applied to your monthly bill. In many cases, that’s enough capacity to run development environments or support a small-scale application, including a personal blog, a portfolio website with contact forms, or even a side project such as a budgeting app or a lightweight CRM handling a few hundred transactions per day.

To learn more about how Aurora DSQL free tier works, visit the AWS Free Tier page.

Additional costs

  • Aurora DSQL automatically replicates data across Availability Zones within a Region to ensure high availability. This inter-AZ replication is included at no additional charge and you do not incur any data transfer costs.

    If you configure a multi-Region cluster, Aurora DSQL automatically replicates writes to each peered cluster in other AWS Regions. These replicated writes incur extra DPU charges equal to the cost of the originating writes. These charges are included in your overall DPU usage, and you do not incur separate data transfer costs.

    Data transferred "in" to and "out" of Aurora DSQL is billed at standard AWS data transfer rates.

      Transfer Type Charges
    Data Transfer IN From Internet & other AWS services No charge
    Data Transfer OUT To AWS Services in same Region No charge
    To AWS Services in different Region Standard AWS data transfer costs apply at source and destination
    To the Internet Standard AWS data transfer costs apply*

    * As part of AWS Free Tier, AWS customers receive 100 GB of data transfer out to the internet free each month, aggregated across all AWS Services and Regions (except China and GovCloud). The 100 GB free tier for data transfer out to the internet is global and does not apply separately or individually to AWS Regions.

  • You can use AWS Backup to back up and restore your Aurora DSQL clusters. AWS Backup is a fully managed service that helps you centralize and automate data protection across AWS services. The following table lists the pricing incurred through AWS Backup.. See AWS Backup pricing for more details.

Pricing examples

  • You run a gaming application with variable activity throughout the day, with traffic peaking in the evenings and on weekends, and dropping significantly during off hours. To support this dynamic workload, you deploy an Aurora DSQL cluster in US East (Ohio).

    Over the course of the month, your Aurora DSQL cluster averages 0.5 DPUs per second, scaling up to 2 DPUs per second during peak times and down to 0.1 DPUs per second during quieter periods. CloudWatch shows the monthly breakdown as:

    • 0.7M WriteDPUs
    • 0.4M ReadDPUs
    • 0.214M ComputeDPUs
    • 1.314M Total DPUs

    In terms of storage, your Aurora DSQL cluster uses 10 GB of storage at the beginning of the month, growing to 20 GB by the end of the month, with an average of 15 GB over the month.

    Note: Aurora DSQL data replication across Availability Zones within a Region (inter-AZ) is included at no additional cost, and you will see no data transfer charges for this replication.

    The estimated total monthly cost for the gaming application would be as follows:

    Usage Calculation Cost
    DPU 1.314M DPU/month * $8/million DPU $10.512
    Storage 15 GB-months * $0.33/GB $4.95
    Data transfer none $ -
    Total bill for the month   $15.462

  • YYou have a mission-critical banking application that processes transactions globally. To ensure high availability and offer low latency access for your users in different geographies, you configure a multi- Region Aurora DSQL setup with one clusters in US East (N. Virginia) and another in US East (Ohio).

    The cluster in Region 1 (N. Virginia) handles all write activity. Over the course of a month, it consumes a total of 6M DPUs. CloudWatch shows the following breakdown:

    • 2M WriteDPUs for processing incoming transactions
    • 2M MultiRegionWriteDPUs to replicate those transactions to the Ohio cluster
    • 0.5M ReadDPUs
    • 1.5M ComputeDPUs
    • 6M Total DPUs

    Meanwhile, the cluster in Region 2 (Ohio) acts as a read replica to serve user requests with low latency. It consumes:

    • 1M ReadDPUs
    • 1.312M ComputeDPUs
    • 2.312M Total DPUs

    In terms of storage, each cluster stores an average of 25 GB of data over the course of the month.

    Note: MultiRegionWrite DPUs reflect the extra work required to replicate data across Regions and are billed in the same Region as the originating writes (in this case, N. Virginia). You won’t see separate data transfer charges for this cross-Region replication as those costs are included in your MultiRegionWriteDPU usage.

    The estimated total monthly cost for your banking application would be as follows:

    Usage Calculation Cost
    DPU in Region 1 1.314M DPUs/month * $8/million DPUs  $48.00
    DPU in Region 2 1.314M DPUs/month * $8/million DPUs  $18.496
    Storage in Region 1 25 GB-months * $0.33/GB $8.25
    Storage in Region 2 25 GB-months * $0.33/GB $8.25
    Data transfer none $ -  
    Total bill for the month   $87.996

FAQs

  • A DPU is the billing unit Aurora DSQL uses to measure the work done by the system to run your SQL workload. This includes both user-initiated SQL statements —such as queries, inserts, updates, and schema changes— as well as related background tasks like statistics updates, index maintenance and auto ANALYZE jobs. Your DPU usage reflects the combined compute and I/O resources consumed by all these activities.

    Aurora DSQL simplifies cost management with DPU-based pricing. Instead of tracking and billing separately for granular infrastructure components like vCPU, memory, network bandwidth, and IOPS, all resource usage is consolidated into a single usage-based unit: the DPU. It reflects the real-time resource consumption of your SQL workload and provides a clear, single line item on your bill. For greater visibility, you can monitor detailed DPU breakdowns (compute, read, write) in CloudWatch to better understand your usage patterns. This helps you stay on top of costs, without the complexity of traditional infrastructure pricing.

  • Aurora DSQL automatically scales compute and storage to match your workload demands, so you only pay for what you use, with no upfront costs. Your monthly bill is charged on two primary components: 1/ database activity, measured in Distributed Processing Units (DPUs) and 2/ storage, measured in GB- month. When your cluster is idle, usage automatically scales to zero, and you incur no DPU charges. Storage costs are based solely on the amount of data you store.

  • Aurora DSQL uses a request-based, usage-driven pricing model. You are billed based on the actual compute and I/O activity generated by your SQL transactions and related background tasks. There are no instances or servers to provision and no hourly charges when the database is idle. This model scales to zero when not in use, so you only pay for what you use.

  • You will find Aurora DSQL pricing familiar if you have experience with other AWS serverless offerings that charge per request, like DynamoDB and Lambda. Similar to DynamoDB on-demand capacity mode, Aurora DSQL bills for reads and writes based on actual usage. However, Aurora DSQL also includes compute charges, like Lambda, based on execution time of SQL expressions and functions. These compute and I/O activities are all measured through a single, unified billing metric called the Distributed Processing Unit (DPU).

  • When using multi-Region clusters with Aurora DSQL, billing is based on the DPUs and storage used per Region. You are charged for the read, write, and compute DPUs in the Region where your request originates. Additionally, writes replicated to peered clusters in other AWS Regions incur extra DPU charges equal to the cost of the originating writes. These charges appear as “MultiRegionWriteDPU” in CloudWatch and are billed in the same Region as the originating writes. You do not incur separate data transfer charges for this replication. For storage, you are charged based on the amount of data stored in each Region. You are not charged for DPUs or storage for your witness Region, as it does not execute transactions or store a data replica.

  • Yes, a detailed breakdown of your DPU usage can be found in the AWS CloudWatch dashboard for your DSQL cluster. This includes four usage metrics:

    1. ComputeDPU – Time spent executing SQL queries
    2. ReadDPU – Resources used to read data from storage
    3. WriteDPU - Resources used to write data to storage
    4. MultiRegionWriteDPU – Resources used to replicate writes to peered clusters in multi-Region configurations.

    When aggregated, these metrics approximate your total monthly DPU usage.

  • No, these metrics are for monitoring purposes only and don’t have separate billing rates. When aggregated, they give you an estimate of your total monthly DPU usage. DPUs are normalized and billed at the same rate whether they are used for compute, read, write, or multi-Region activity. For example, 1M DPUs cost $8 in US East (Ohio), whether that’s 400K compute, 300K write, and 300K read, or 300K write and 700K read. The mix doesn’t affect your total cost.

  • With Aurora DSQL free tier, the first 100,000 DPUs and 1 GB of storage is billed at no charge each month. Any excess DPUs and storage beyond the free tier amount in a given month will be charged at the Regional DPU and GB-month storage cost. For example, if you use 150,000 DPUs and 5 GB of storage last month, you would be charged for 50,000 DPUs and 4 GB of storage. If you have an organization set up through AWS Organizations, this applies per management account.

  • Your results may vary, but to establish a reference for what can be accomplished with 100K DPUs, we executed a small benchmark with a 95/5 read/write mix using transactions derived from the TPC-C benchmark. Based on this benchmark, 100K DPUs were equivalent to ~700,000 TPC-C transactions.