Cloud commitment models have been evolving steadily over the last few years. For compute workloads, flexible, spend-based commitments made it easier for teams to reduce costs without locking themselves into rigid infrastructure decisions.
Databases, however, didn’t benefit from the same flexibility.
Until recently, most managed database services required Reserved Instances, commitments tied to specific instance types, engines, and Regions. For teams running modern database environments that scale frequently, resize often, or change deployment models, this made long-term cost optimization difficult and, at times, risky.
That’s beginning to change now.
With the introduction of AWS Database Savings Plans in December last year, database commitments are finally moving toward a spend-based model. Instead of planning around fixed configurations, teams can now commit to a consistent level of database spend and have discounts applied automatically across eligible usage.
Today, we’re sharing our announcement on how Usage.ai now supports this new commitment model and what that means for teams optimizing managed database workloads.
As of January 20th, 2026, Database Savings Plans coverage will be available in Usage.ai for the following AWS managed database services:
● Amazon RDS: Supported for newer-generation RDS and Aurora instance families (Gen 7 and newer, such as db.m7, db.r7, and equivalent Graviton-based families). Older RDS families (including M5, R5, T3, and T4g) are not eligible for Database Savings Plans and require Reserved Instances or migration to newer families to benefit from spend-based discounts.
● Amazon ElastiCache: Supported only for ElastiCache for Valkey (Redis-compatible), including both provisioned clusters and serverless deployments. Standard Redis and Memcached engines are excluded and continue to require Reserved Nodes for commitment-based savings.
● Amazon Neptune: Supported for both provisioned and serverless Neptune deployments running on newer-generation instance families (Gen 7 and above, including modern Graviton-based families such as r7g and m7g), enabling spend-based savings as graph workloads scale.
● Amazon DocumentDB: Supported for DocumentDB instance-based and serverless usage on newer-generation instance families (such as M7 and R7) across Single-AZ, Multi-AZ, and serverless deployment models, without locking commitments to specific instance sizes.
With this release, our customers can:
● Model spend-based database commitments instead of instance-specific Reserved Instances
● Evaluate coverage across dynamically resizing and multi-engine database environments
● Reduce manual effort tied to RI planning, tracking, and renewals
● Confidently align long-term commitments with workloads that change over time
Historically, database commitments were one of the hardest areas to automate.
Reserved Instances required teams to predict, often months in advance:
● Instance class
● Engine
● Deployment model
● Region
For workloads running on RDS, ElastiCache, or Neptune, even routine changes like resizing, scaling, or upgrading instance generations could break commitment coverage and lead to wasted spend.
Database Savings Plans change that equation. Because commitments are spend-based, not configuration-based, Usage.ai can now apply the same automated coverage modeling used for compute to managed databases, without locking customers into fragile assumptions.
This is especially valuable for teams operating:
● Multi-engine database environments
● Frequently scaling or resizing workloads
● Modern architectures that evolve faster than traditional RI planning cycles
From a product perspective, this release expands the commitment surface area that Usage.ai can optimize. Inside Usage.ai, Database Savings Plans are evaluated alongside existing commitments to:
● Determine optimal hourly spend levels
● Maximize coverage while minimizing overcommitment risk
● Support cleaner transitions at RI expiration windows
Rather than managing database commitments as static, isolated assets, Usage.ai treats them as part of a unified optimization strategy, continuously aligned with real usage patterns.
In the next rollout cycle, Usage.ai will extend Database Savings Plans support to additional AWS services, including:
● Amazon DynamoDB Keyspaces
● Amazon Timestream
While there is no specific timeline to share yet, they are actively on our roadmap as AWS continues to expand Database Savings Plans coverage.
AWS’s move toward spend-based commitments for databases reflects a broader shift in how cloud infrastructure is built and operated. Databases are no longer static, single-configuration systems and commitment models are finally catching up to that reality.
With today’s update, Usage.ai customers can begin applying Database Savings Plans to more of their managed database footprint, with the same confidence and automation they expect from compute optimization.
We’ll continue to roll out support as AWS expands the scope of Database Savings Plans and we’ll share updates as new services become available.
To explore database commitment optimization in Usage.ai, visit usage.ai.
Share this post
.png)
