AWS, Database Savings Plans

AWS Database Savings Plans: pricing, coverage, and buying guide for 2026

Understand AWS Database Savings Plans eligibility, pricing, and commitment sizing so you can capture eligible savings without overcommitting your database spend.

Priyanka

Content Strategist

Updated August 19, 2026

17 min read

Key takeaways

AWS Database Savings Plans use a $/hour commitment rather than a commitment to one database instance.

The plan has a one-year term with no upfront payment.

AWS publishes savings of up to 35%. Actual savings depend on eligible usage, rates, and utilization.

Database Savings Plans apply to eligible serverless usage and latest-generation provisioned instances.

They are available in AWS Regions except China Regions.

AWS explicitly prevents a Database Savings Plan from stacking with an RDS Reserved Instance or DynamoDB reserved capacity on the same workload.

Size a commitment against a durable hourly floor, not peak or average database spend.

Understand AWS Database Savings Plans eligibility, pricing, and commitment sizing so you can capture eligible savings without overcommitting your database spend.

AWS Database Savings Plans reduce eligible database costs by up to 35% in exchange for a consistent hourly usage commitment over a one-year term with no upfront payment. Available in AWS Regions outside China, they apply automatically to eligible latest-generation provisioned instances and support serverless usage across 10 AWS database and data services.

For FinOps teams, the main advantage is flexibility. Instead of committing to one fixed database configuration, the discount can follow eligible usage as workloads move across supported engines, instance families, sizes, deployment options, and regions.

What are AWS Database Savings Plans?

AWS Database Savings Plans are a spend-based commitment model for managed database usage. You commit to a consistent Savings Plans spend rate per hour, and AWS automatically applies discounted pricing to eligible usage until the hourly commitment is consumed.

Usage above the commitment is charged at applicable On-Demand rates.

Unlike a service-specific reservation, the commitment can follow eligible usage as configurations change. As explained in the AWS Database Savings Plans launch announcement, AWS gives examples such as changing Aurora instance types or Regions, moving from RDS for Oracle to Aurora PostgreSQL, or shifting eligible database usage between supported services.

For the broader commitment model, see our AWS Savings Plans guide.

Which services do AWS Database Savings Plans cover?

According to the AWS Savings Plans types documentation, Database Savings Plans currently cover 10 supported services.

Service	Eligible Usage to Know About
Amazon RDS	Latest-generation eligible provisioned instances
Amazon Aurora	Latest-generation instances, Aurora Serverless v2, and Aurora DSQL
Amazon DynamoDB	On-demand and provisioned throughput
Amazon ElastiCache	Valkey instances and ElastiCache Serverless for Valkey
Amazon DocumentDB	Eligible provisioned instances and DocumentDB Serverless
Amazon Neptune	Eligible instances, Neptune Serverless, and Neptune Analytics
Amazon Keyspaces	Eligible on-demand and provisioned throughput
Amazon Timestream	Eligible Timestream usage
AWS DMS	Latest-generation provisioned instances and DMS Serverless
Amazon OpenSearch Service	Eligible provisioned and serverless usage

Eligibility can expand as AWS introduces new instance generations and offerings. Verify the current AWS Database Savings Plans pricing page before purchasing.

AWS states that serverless deployments can provide up to 35% savings, while provisioned instances across supported database services can provide up to 20%. DynamoDB and Keyspaces differ, with up to 18% for on-demand throughput and 12% for provisioned throughput.

These are AWS-published maximum discounts, not guaranteed realized savings.

What is not covered?

A few exclusions matter before calculating potential savings:

Database Savings Plans are not available in AWS China Regions.
Older provisioned instance generations may fall outside DSP eligibility because AWS scopes provisioned coverage to the latest generations.
For RDS for SQL Server, DSP discounts apply only to the instance price. Windows Server and SQL Server licensing charges remain at On-Demand rates.
Amazon Redshift is not among the 10 services AWS currently lists for Database Savings Plans.
An RDS Reserved Instance or DynamoDB reserved capacity discount cannot stack with DSP on the same workload.

How AWS Database Savings Plans work

Every supported usage type has an On-Demand rate and a Savings Plans rate. Eligible usage consumes the hourly commitment at the discounted Savings Plans rate. Usage beyond that commitment remains at On-Demand pricing.

Useful commitment formulas

Annual commitment obligation

Hourly commitment × 24 × 365

At $4/hour

$4 × 8,760 = $35,040 per year

For a simplified workload with a 20% discount:

Approximate On-Demand-equivalent coverage

Hourly commitment ÷ (1 − discount rate)

At $4/hour and 20% discount

$4 ÷ 0.80 = $5/hour

A $4/hour Savings Plans commitment could therefore cover approximately $5/hour of equivalent On-Demand usage when the applicable discount is exactly 20%.

This is an illustrative estimate. Actual Database Savings Plans rates vary by service and usage type.

Database Savings Plans vs Reserved Instances

Factor	Database Savings Plans	Reserved Instances
Commitment basis	Hourly spend	Matching resource attributes
DSP term	1 year	Typically 1 or 3 years where offered
DSP payment structure	No upfront payment	Varies by offering
Cross-Region flexibility	Yes for eligible DSP usage	More restricted
Cross-service flexibility	Across DSP-supported services	No
Serverless support	Supported for eligible DSP services	Not instance-reservation based
Best fit	Changing database estates	Stable, predictable configurations

Database Savings Plans can suit workloads that may change database services, instance types, Regions, or deployment models during the commitment.

Reserved Instances can remain attractive for predictable workloads when a service-specific reservation produces stronger economics.

For more detail, see our RDS Reserved Instances guide and ElastiCache Reserved Nodes guide.

Pre-purchase eligibility check

Before sizing an AWS Database Savings Plan, confirm:

Your workloads run outside AWS China Regions.

The service appears on the current DSP eligibility list.

Provisioned workloads use an eligible latest-generation instance.

You have separated non-discountable charges such as SQL Server licensing.

Existing RDS RIs or DynamoDB reserved capacity are removed from the same-workload opportunity.

Expected eligible usage should remain through the one-year commitment.

Always verify current eligibility and rates before purchasing.

How to size a Database Savings Plan step by step

Isolate eligible hourly spend

Separate DSP-eligible database usage from storage, licensing, unsupported generations, and other excluded charges.

Find the durable usage floor

Look for the hourly level that persists during nights, weekends, and low-demand periods. Do not automatically commit against average or peak spend.

Remove expected reductions

Exclude workloads scheduled for decommissioning, rightsizing, or major architectural changes that could reduce eligible usage.

Model several commitment levels

AWS provides Savings Plans Recommendations and Purchase Analyzer for estimating savings, utilization, and coverage across custom purchase scenarios.

Treat recommendation outputs as estimates rather than guarantees.

Purchase and monitor

Review the hourly commitment in AWS Billing and Cost Management, purchase the plan after validating the economics, and track performance through Savings Plans coverage and utilization reports.

Practical callout: commit to the floor, not the forecast.

A $20/hour monthly average can hide repeated periods at $13/hour. If those lows are normal rather than temporary, sizing against the average can create commitment exposure when demand falls.

Practical Database Savings Plan example

Assume eligible RDS and Aurora usage ranges from $14/hour during quiet periods to $24/hour during peaks, averaging $19/hour.

Instead of starting with the $19/hour average, evaluate the persistent $14/hour baseline first.

Then convert the actual eligible usage mix to Savings Plans rates using AWS Purchase Analyzer. A single assumed discount percentage should only be used for rough modeling because AWS rates vary across services and usage types.

For service-level analysis, see our DynamoDB on-demand pricing guide and Neptune Serverless pricing guide.

How we approach Database Savings Plans

For Database Savings Plans, use AWS Savings Plans Recommendations and Purchase Analyzer to evaluate eligible spend, coverage, utilization, and custom commitment-sizing scenarios before purchasing.

Commitments purchased from approved recommendations and managed through our Flex Commitment Program are identified as Flex Commitments.

Cashback protection is tied to an actual loss threshold. A loss occurs when the cost of a Flex Commitment exceeds what the same usage would have cost at the On-Demand rate. We calculate applicable losses monthly and provide cashback according to the Flex Commitment program process and terms.

This protection does not change the AWS Database Savings Plan itself. AWS DSP remains a one-year commitment with AWS-published savings of up to 35%.

Where Flex Commitments fit beyond database spend

AWS database and compute commitments should not be conflated.

Database Savings Plans cover eligible database usage with a one-year term and up to 35% AWS-published savings. AWS Compute Savings Plans and EC2 Instance Savings Plans are separate commitment products for compute workloads, with their own terms and discount levels.

For eligible AWS commitment opportunities outside Database Savings Plans, With Flex Insured Commitments, teams can get up to 57% savings of a three-year AWS commitment with none of the commitment risk, depending on the service, configuration, and payment option.

If a commitment ever costs more than the equivalent on-demand usage, we calculate that loss at the end of the month and provide cashback protection for it.

EVALUATE YOUR AWS DATABASE SAVINGS

See what your database commitment covers.

Review your database baseline, coverage, and utilization risk before committing.

Book a savings review

Frequently asked questions

How much can AWS Database Savings Plans save?

AWS publishes savings of up to 35%. Actual savings depend on the service, usage type, applicable Savings Plans rate, and how effectively the commitment is utilized.

Where are AWS Database Savings Plans available?

Database Savings Plans are available in AWS Regions outside China Regions. Individual services must also be available in the Region where you intend to use them.

Does AWS DMS qualify for Database Savings Plans?

Yes. AWS lists DMS among supported services. DSP applies to latest-generation provisioned instances and DMS Serverless, subject to current eligibility.

Can Database Savings Plans and Reserved Instances be combined?

They can be used for different workloads. AWS specifically states that Database Savings Plans cannot be combined with an RDS Reserved Instance or DynamoDB reserved capacity discount on the same workload.

Should I choose Database Savings Plans or Reserved Instances?

Database Savings Plans generally fit teams that value flexibility across eligible services, Regions, instance types, and deployment models.

Reserved Instances can remain attractive for highly stable workloads when the applicable reservation offers better economics. Compare both using actual usage and current AWS prices before purchasing.