New York, USA — June 16, 2026 — Usage.ai, a cloud cost optimization platform specializing in automated AWS commitment management, has announced full automation support for AWS Database Savings Plans (DSP) across all 10 AWS managed database services currently eligible under the program. The announcement follows AWS's March 2026 expansion of DSP eligibility to include OpenSearch Service and Neptune Analytics.
Usage.ai customers can now automate the complete DSP lifecycle, spend analysis, commitment sizing, purchase, utilization monitoring, and cashback on underutilized commitments for every service covered under AWS Database Savings Plans.
What the Announcement Covers
Usage.ai's Database Savings Plans automation is now live across:
- Amazon RDS: Gen 7+ provisioned instances (db.r7, db.r8g, db.m7, db.m7g families)
- Amazon Aurora: Gen 7+ provisioned instances, Aurora Serverless v2, and Aurora DSQL
- Amazon DynamoDB: on-demand throughput (up to 18% savings) and provisioned capacity (up to 12% savings)
- Amazon ElastiCache: Valkey engine only, covering Gen 7+ provisioned clusters and ElastiCache Serverless for Valkey
- Amazon DocumentDB: Gen 7+ provisioned instances and DocumentDB Serverless
- Amazon Neptune: Gen 7+ provisioned instances, Neptune Serverless, and Neptune Analytics (added by AWS on March 5, 2026)
- Amazon Keyspaces: on-demand throughput (up to 18% savings) and provisioned throughput (up to 12% savings); DSP is the only commitment discount path for this service, as no Reserved Instances are available
- Amazon Timestream: Timestream for InfluxDB instances
- Amazon OpenSearch Service: Serverless and Gen 7+ provisioned instances (added by AWS on March 5, 2026)
- AWS Database Migration Service (DMS): Gen 7+ replication instances and DMS Serverless
How Usage.ai Handles the Full DSP Lifecycle
Purchasing a Database Savings Plan at the right commitment level requires knowing the consistent floor of eligible hourly spend, the amount eligible database costs never drop below across all hours. Over-committing results in paying for capacity that goes unused; under-committing leaves savings unrealized. Managing this across 10 services with different eligibility rules is operationally expensive to do manually.
Usage.ai automates the process in five stages:
Analysis: The platform pulls Cost and Usage Report (CUR) data and identifies DSP-eligible spend across all 10 services, separating it from RI-eligible spend to avoid double-counting.
Commitment sizing: Usage.ai calculates the consistent hourly floor spend on DSP-eligible usage across the prior 60 days to determine the correct commitment level.
Purchase: The DSP commitment is purchased through billing-layer access and activates immediately.
Monitoring: DSP utilization is monitored on a 24-hour refresh cycle, compared to AWS Cost Explorer's 72-hour-plus refresh window.
Cashback on underutilization: If a DSP commitment becomes underutilized due to decommissioning, migration, or downsizing, Usage.ai provides cashback on the unused committed amount. This applies to DSP commitments on the same terms as Savings Plans and Reserved Instances across the rest of the platform.
Usage.ai's fee structure is a percentage of realized savings only.
Recommended Order of Operations
Usage.ai advises teams to follow a specific sequence before purchasing DSP commitments:
- Right-size instances based on actual CPU and memory utilization.
- Migrate to current-generation instances: Gen 7+ for RDS and Aurora; Valkey engine for ElastiCache, since older-generation instances remain ineligible for DSP and continue to require Reserved Instances.
- Evaluate storage configuration, particularly for Aurora and DocumentDB where the choice between Standard and I/O-Optimized storage carries meaningfully different cost implications depending on actual I/O consumption.
- Purchase DSP commitments on the confirmed, right-sized, current-generation spend floor.
Teams that purchase DSP before completing these steps lock in commitments on oversized or ineligible infrastructure, discounting a spend level higher than necessary.
Key Facts About AWS Database Savings Plans
- Term: 1-year only (no 3-year option)
- Payment: No Upfront only (no All Upfront or Partial Upfront options, unlike Compute Savings Plans)
- Maximum savings: Up to 35% for serverless workloads (Aurora Serverless v2, Neptune Serverless, DocumentDB Serverless, ElastiCache for Valkey Serverless); up to 20% for most provisioned workloads; up to 18% for on-demand throughput workloads (DynamoDB, Keyspaces); up to 12% for DynamoDB and Keyspaces provisioned capacity
- Application order: DSP applies after Reserved Instances in the discount waterfall and cannot be combined with RIs or reserved capacity on the same workload in the same billing hour
- DynamoDB note: DSP cannot be combined with DynamoDB reserved capacity on the same table
- ElastiCache note: Only the Valkey engine is covered; standard Redis OSS and Memcached continue to require Reserved Nodes
Leadership Perspective
Kaveh Khorram, CEO of Usage.ai, framed the broader context: "FinOps started with EC2. It matured into Savings Plans and Reserved Instances across compute. The next phase is database and it's more complex because the eligibility rules, the instance families, and the discount waterfall all behave differently across 10 services. The teams that get this right balance engineering judgment with financial discipline, and that balance is where cloud economics are won or lost."
Additional Resources
As AWS continues to expand Database Savings Plans eligibility, teams managing database workloads at scale will need to account for service-specific eligibility rules, instance generation requirements, and the correct sequencing of commitments alongside existing Reserved Instances. Full coverage details, discount rates, and service-specific eligibility information are available in the AWS Database Savings Plans: Complete Guide for 2026. Teams can review their DSP-eligible spend at usage.ai/savings-calculator. The complete announcement is at usage.ai/news.
About Usage.ai
Usage.ai is a cloud cost optimization platform that helps engineering and finance teams cut AWS, Azure, and GCP spend by 30 to 50 percent without infrastructure changes, long-term contracts, or stranded commitments. Its Insured Commitments model provides cashback and credit guarantees on every dollar of unused capacity. Usage.ai is SOC 2 Type II certified and headquartered in New York City.
Top comments (0)