RDS Rightsizing Tools and RDS Cost Optimization Software Compared With Prices
Amazon RDS is usually the second largest line on an AWS bill after EC2, and it is harder to shrink. Changing a DB instance class causes an outage, allocated storage cannot simply be reduced, Multi-AZ doubles every mistake, and since 2024 a database left on an old engine version pays an Extended Support surcharge per vCPU on top of everything else. RDS rightsizing tools split the same way EC2 tools do: most recommend a smaller instance class and leave the change to your DBA, and a few will make the change for you within guardrails. This page compares twelve options for RDS and Aurora, says plainly which ones recommend and which ones resize, lists the RDS catch for each, and gives every price that is published, including AWS Marketplace rate cards most vendors leave off their own sites.
projected this month if unattended
Spend by team
Budget forecast
The short answer
The best RDS rightsizing tool for most AWS teams is AWS Compute Optimizer, because its RDS and Aurora recommendations are free and cover instance class, storage, Graviton and Aurora I/O-Optimized. It supports only MySQL, PostgreSQL and Aurora, reads memory only through Performance Insights, and never applies a change. Buy IBM Turbonomic or Sedai when you want RDS resized for you, Datadog or Vantage when you want the recommendations next to your other cost data, and fix Extended Support and storage before you buy anything.
CostAnalyst is on this list, so here is the plain version first. We do not resize databases and we hold no access to your AWS account. We read the AWS billing export you upload, show RDS and Aurora spend by account, team, engine and instance family, flag idle and oversized spend, Extended Support charges and anomalies, and show on the next bill whether a resize saved money. We publish our prices. Everything below about other tools comes from AWS pricing pages, the AWS Price List and documentation, or from each vendor's own documentation and AWS Marketplace listing, read on 6 October 2026.
How we compared
Five things that actually separate these tools
Does the tool recommend the new instance class, or resize the database for you
Settle this first, because a database resize is not like an EC2 resize. AWS documents that modifying the DB instance class causes an outage during the change, applied in the next maintenance window unless you apply it immediately, and Multi-AZ only shortens it through a failover. Recommendation tools need CloudWatch and billing access and clear security review quickly. Tools that resize need permission to modify your production databases and a change process your DBAs have signed, which is why most teams start with recommendations even when they plan to automate later.
What AWS gives you for free, and where it stops
AWS Compute Optimizer recommends RDS for MySQL, RDS for PostgreSQL and Aurora instance classes, storage type, IOPS and throughput, Graviton alternatives and the Aurora Standard versus I/O-Optimized choice, and flags idle databases, all at no charge from 14 or 32 days of metrics. Enhanced infrastructure metrics extend the lookback to 93 days at $0.0003360215 per resource-hour, about $0.25 per database a month. The limits: no recommendations for MariaDB, SQL Server, Oracle or Db2, memory is read only through Performance Insights, and its Automation feature applies EBS actions only, never an RDS change. Cost Optimization Hub collects the same RDS findings free, and Trusted Advisor shows them as checks only on Business Support+ or higher.
Can it see memory, the metric that decides whether a downsize is safe
CPU is rarely what breaks after a database downsize. Buffer cache and working memory are. Compute Optimizer reads CPU, network, read and write operations, connections and free storage from CloudWatch, and uses memory only through Performance Insights counters such as swap activity on RDS and out-of-memory kills on Aurora. A tool that ignores memory will recommend dropping a cache-heavy PostgreSQL instance from a db.r6i.2xlarge to an xlarge because CPU averages 15 percent, and the first busy Monday will show why that was wrong. Ask every vendor which memory signal it uses.
What the paid tools cost, and in which unit
RDS tools price in four units, which is why quotes are hard to compare. IBM Turbonomic sells per managed virtual server, $37,909.82 a year for 200 on AWS Marketplace with $22.60 for each one above that. Kubex sells vCPU tiers, $499 a month for 500 vCPUs, though its listing does not say how RDS counts against them. Datadog Cloud Cost Management charges $5 per $1,000 of monitored cloud spend on Pro, and Vantage charges a flat $30 or $200 a month by tracked spend. Sedai, nOps, CAST AI Database Optimizer, CloudHealth recommendations and Cloudability sell on quotes or spend bands, and commitment tools such as Usage.ai take a share of savings, 35 percent on RDS on its Marketplace listing.
Will the saving reach the invoice, and what to fix before you resize
Three things eat RDS savings before a rightsizing tool does. Reserved Instances are size flexible within a family for MySQL, MariaDB, PostgreSQL, Oracle BYOL and Aurora, so a downsize frees commitment that sits unused until something else fills it, and Database Savings Plans, launched in December 2025, are a one-year hourly commitment that a downsize does not lower. Extended Support costs $0.10 per vCPU-hour in years one and two and $0.20 in year three, doubled for Multi-AZ, and MySQL 8.0 standard support ended on 31 July 2026. And allocated storage can only grow: shrinking it means a Blue/Green deployment, available since November 2024 for PostgreSQL 12 and later, MySQL 5.7 and later and MariaDB 10.4 and later. Measure on the bill, not in the tool.
At a glance
12 RDS rightsizing tools compared
← Scroll to see all columns →
| Tool | Best for | Recommends or resizes | The RDS catch | Pricing |
|---|---|---|---|---|
| AWS Compute Optimizer | Every AWS account running MySQL, PostgreSQL or Aurora, as the free baseline | Recommends instance class, storage, Graviton and Aurora storage type; flags idle; applies nothing to RDS | MySQL, PostgreSQL and Aurora only; memory only through Performance Insights | Free; enhanced infrastructure metrics $0.0003360215 per resource-hour, about $0.25 per database a month |
| Cost Optimization Hub and Trusted Advisor | FinOps teams who want RDS savings ranked in the Billing console | Ranks RDS instance, storage, Aurora storage and RDS Reserved Instance actions; changes nothing | Trusted Advisor shows these checks only on Business Support+ or higher | Cost Optimization Hub free; Trusted Advisor full checks from $29 per account a month on Business Support+ |
| IBM Turbonomic | Large estates that want RDS scaling and parking run as policy, across every engine | Generates RI-aware scale and park actions for RDS compute and storage across seven engines | Sold per managed virtual server, so a small RDS fleet still buys the 200-server block | $37,909.82 a year for 200 managed virtual servers on AWS Marketplace, $22.60 per extra server |
| Sedai | Teams that want RDS and Aurora resized autonomously within guardrails | Applies RDS and Aurora rightsizing, storage and I/O changes within guardrails you set | No public price; the AWS listing offers only a 30-day proof of concept | Quote only; AWS Marketplace lists a 30-day POC, paid tiers are private offers |
| Kubex (formerly Densify) | Teams that want RDS sizing in the same tool as Kubernetes and EC2 | Classifies RDS as under or over-provisioned and recommends class, IOPS and RIs; automation optional | The listing prices vCPU tiers and does not say how RDS counts against them | $499 a month or $6,000 a year to 500 vCPUs; $900 a month or $10,800 a year to 1,000 vCPUs |
| nOps | AWS-only teams that want RDS recommendations plus automated commitments | Recommends RDS resizing and gp2 to gp3; automates commitment purchases, not resizing | No public price; the AWS listing shows only placeholder dimensions | Fixed fee by cloud spend for visibility, share of savings for rate optimization; 14-day trial |
| Datadog Cloud Cost Management | Teams already on Datadog that want RDS cost actions next to database metrics | Recommends RDS downsizing, PIOPS cuts, Graviton and Extended Support migrations; can delete idle RDS | Priced on all monitored cloud spend, not on the databases you rightsize | Pro $5 per $1,000 of monitored spend a month billed annually; Enterprise $10 per $1,000 |
| CAST AI Database Optimizer | Read-heavy PostgreSQL or MySQL on RDS and Aurora that could downsize behind a cache | Adds a caching proxy that offloads reads so you can downsize; does not resize RDS itself | A different architecture, not a rightsizing tool; price is quote only | Quote only, contact form on the CAST AI pricing page |
| Vantage | Smaller teams that want RDS rightsizing and Extended Support alerts at a flat price | Surfaces RDS rightsizing, idle and Extended Support recommendations; changes nothing on RDS | Builds on Compute Optimizer findings, so its RDS engine coverage matches AWS | Free to $2,500 of monthly tracked spend; Pro $30 and Business $200 a month; Enterprise custom |
| CloudHealth (Broadcom) | Enterprises that want RDS rightsizing inside a full multi-cloud cost platform | RDS rightsizing report with projected price and savings; changes nothing | Instance must run 48 hours to qualify, and Aurora Serverless is excluded | $45,000 a year to $150,000 of monthly AWS spend on AWS Marketplace, $90,000 to $300,000 |
| IBM Cloudability | Finance-led FinOps teams already in the IBM Apptio stack | Rightsize and terminate recommendations for RDS on on-demand or effective cost; changes nothing | Priced on annual cloud spend tiers that jump at $1 million and $3 million | $30,000 a year to $1M of annual spend, $76,680 to $3M, on AWS Marketplace |
| CostAnalyst | Finance and FinOps teams proving what RDS rightsizing saved on the AWS bill | Flags idle, oversized and Extended Support RDS spend from the bill; changes nothing | Works from the billing export you upload, not from CloudWatch or Performance Insights | $59, $179 and $479 a month billed yearly, published |
Vendors change pricing and packaging often, so confirm the current figure before you buy.
Tool by tool
What each tool is genuinely best at
AWS Compute Optimizer
Best for: Every AWS account running MySQL, PostgreSQL or Aurora, as the free baseline
Compute Optimizer is the tool to switch on first, in the management account for the whole organization. For RDS and Aurora it classifies each DB instance as over-provisioned, under-provisioned or optimized, recommends a new instance class and a storage configuration with volume type, IOPS and throughput, shows the Graviton price if you set the architecture preference, and tells Aurora clusters whether Standard or I/O-Optimized storage would be cheaper. Idle databases, meaning no connections and low CPU and I/O over the lookback, get a stop or snapshot-and-delete suggestion, and idle Aurora instances a move to db.serverless. It needs 30 hours of metrics and refreshes daily. It never changes a database, and it says nothing about MariaDB, SQL Server or Oracle.
See what AWS Compute Optimizer costsCost Optimization Hub and Trusted Advisor
Best for: FinOps teams who want RDS savings ranked in the Billing console
Cost Optimization Hub pulls Compute Optimizer RDS recommendations together with RDS Reserved Instance purchase recommendations and Aurora storage findings, deduplicates them and estimates savings after your discounts, at no extra charge. Trusted Advisor exposes the same data as four RDS cost checks for DB instances, DB instance storage, Aurora cluster storage and Reserved Instance purchases, and its old Idle DB Instances check is now legacy. The hub is the right place for finance to see the size of the RDS opportunity. Neither will resize anything, and neither allocates the saving to the team that owns the database.
See what Cost Optimization Hub and Trusted Advisor costsIBM Turbonomic
Best for: Large estates that want RDS scaling and parking run as policy, across every engine
Turbonomic treats each RDS instance as a database server it can scale up or down in compute and storage, park on a schedule, and size with your Reserved Instances in mind, and it covers more engines than AWS does: Aurora MySQL and PostgreSQL, MySQL, PostgreSQL, MariaDB, SQL Server and Oracle. IBM describes the platform as executing safe, policy-driven actions, but its RDS documentation does not say whether database scale actions can run in automatic mode, so confirm that in the trial before you count on it. Its public AWS Marketplace contract covers 200 managed virtual servers for a year, with 24 and 36 month terms at straight multiples. It earns its price when EC2, RDS and Kubernetes are optimized from one policy engine, not for a dozen databases.
IBM Turbonomic compared to CostAnalystSedai
Best for: Teams that want RDS and Aurora resized autonomously within guardrails
Sedai sells autonomous optimization and lists databases as a first-class target: it rightsizes RDS and Aurora instances and tunes storage and I/O, applying changes itself within guardrails you define. That makes it the most direct answer to a backlog of RDS recommendations nobody has time to apply. The catch is buying it: Sedai publishes no price on its site, and its AWS Marketplace listing currently offers only a 30-day proof of concept, so expect a sales process and a private offer.
Sedai compared to CostAnalystKubex (formerly Densify)
Best for: Teams that want RDS sizing in the same tool as Kubernetes and EC2
Kubex, the current name for Densify, reads RDS CloudWatch metrics, classifies each instance as under or over-provisioned, and recommends an instance class, IOPS sizing and Reserved Instance changes, next to its better known Kubernetes and EC2 analysis. Kubex says implementation can be automated, but its RDS page gives no detail on automated database changes, so treat it as recommendation first. It publishes the clearest price of any paid tool here, two auto-renewing AWS Marketplace tiers. Ask before signing whether your RDS vCPUs count toward the tier.
Kubex (formerly Densify) compared to CostAnalystnOps
Best for: AWS-only teams that want RDS recommendations plus automated commitments
nOps flags underutilized RDS instances and storage upgrades inside its AWS cost platform, and its stronger product is commitment management, which it does buy and sell for you. For RDS rightsizing it is a recommender: your DBA still makes the change. Pricing is a fixed fee based on cloud spend for cost visibility and a percentage of realized savings for rate optimization, with a 14-day free trial. Its AWS Marketplace listing carries only placeholder dimensions, so there is no list price to compare.
See what nOps costsDatadog Cloud Cost Management
Best for: Teams already on Datadog that want RDS cost actions next to database metrics
Datadog turns Compute Optimizer and its own analysis into RDS recommendations: delete unused instances, downsize instances, cut provisioned IOPS below 80 percent use over two weeks, migrate to Graviton, and migrate engines that are paying Extended Support. Since DASH 2025 you can terminate an idle RDS instance from Cloud Cost Management, while downsizing stays a recommendation. It is strongest when the same team already watches database performance in Datadog, because the cost finding and the latency graph sit on one screen. The price scales with your whole cloud bill.
See what Datadog Cloud Cost Management costsCAST AI Database Optimizer
Best for: Read-heavy PostgreSQL or MySQL on RDS and Aurora that could downsize behind a cache
CAST AI DBO takes another route to a smaller database bill. It discovers RDS and Aurora instances and places an autonomous caching layer in front of PostgreSQL and MySQL, so repeated reads stop reaching the database and the instance can then be downsized. CAST AI says most customers see at least 30 percent lower database infrastructure cost after downsizing. It changes your data path, which means application testing, and it has no published price, unlike the CAST AI Kubernetes product whose rate card is public on AWS Marketplace.
See what CAST AI Database Optimizer costsVantage
Best for: Smaller teams that want RDS rightsizing and Extended Support alerts at a flat price
Vantage shows Compute Optimizer RDS rightsizing and idle recommendations inside its cost reports, and adds notices for databases approaching end of standard support so Extended Support charges do not arrive as a surprise. Its automation is aimed at Savings Plans, not at databases. The attraction is price: tiers are flat and published by tracked spend, and you can evaluate it end to end without a sales call.
Vantage compared to CostAnalystCloudHealth (Broadcom)
Best for: Enterprises that want RDS rightsizing inside a full multi-cloud cost platform
CloudHealth includes an RDS rightsizing dashboard that lists each instance with its current and projected monthly price and the saving, within the same platform that handles allocation, budgets and commitment planning. It is recommendation only, instances need 48 hours of activity, and Aurora Serverless is out of scope. As a standalone RDS tool it is expensive. As part of a platform you were buying anyway, the RDS report comes with it.
See what CloudHealth (Broadcom) costsIBM Cloudability
Best for: Finance-led FinOps teams already in the IBM Apptio stack
Cloudability has an AWS RDS rightsizing view with rightsize and terminate recommendations, where idle is judged by database connections, and it can show savings on on-demand or effective cost, which matters when Reserved Instances cover the instance. Like CloudHealth it is bought as a FinOps platform and the RDS view comes with it.
See what IBM Cloudability costsCostAnalyst
Best for: Finance and FinOps teams proving what RDS rightsizing saved on the AWS bill
CostAnalyst reads the AWS billing export you upload and breaks RDS and Aurora spend down by account, team, engine and instance family next to your other cloud and SaaS spend. It flags idle and oversized database spend, Extended Support line items and anomalies, and shows on the next bill whether a resize or a Database Savings Plan actually lowered the invoice. Pair it with Compute Optimizer for the instance class recommendation, or with one of the resizing tools above if you want the change made for you.
See how CostAnalyst worksHow to choose
Pick by the problem you actually have
You run a few dozen MySQL or PostgreSQL databases and DBAs apply changes by hand
Turn on Compute Optimizer and Cost Optimization Hub in the management account, add Performance Insights where memory matters, and work the list in maintenance windows. Spend nothing on tooling until the backlog outgrows the team.
You have hundreds of databases across several engines, including SQL Server or Oracle
AWS will not recommend for those engines. IBM Turbonomic generates actions for all of them, and its 200-server contract makes sense at this size.
You want the changes applied for you and can accept a sales process
Shortlist Sedai and Turbonomic, run both on non-production databases first, and compare what each changed against the bill a month later.
Your team already lives in Datadog
Use Datadog Cloud Cost Management so RDS cost findings sit next to query latency, and check what $5 per $1,000 of all monitored spend comes to before you sign.
The databases run old engine versions
Fix Extended Support before rightsizing. At $0.10 to $0.20 per vCPU-hour, doubled on Multi-AZ, an upgrade often saves more than a downsize, and Vantage or Datadog will flag the affected instances.
Finance wants the savings proven by someone other than the vendor
Keep the measurement outside the tool that makes the change. Allocate RDS spend by team from the AWS bill before the rollout and compare the same view after it. CostAnalyst does that from the export you upload, at a published price.
Questions buyers ask
RDS rightsizing tools, answered
What is the best RDS rightsizing tool?
For most teams, AWS Compute Optimizer, because it recommends RDS and Aurora instance classes, storage, Graviton moves and the Aurora storage type for free. It covers only MySQL, PostgreSQL and Aurora and never applies a change. When you need other engines or automated changes, IBM Turbonomic and Sedai are the main paid options.
Does AWS Compute Optimizer support RDS?
Yes, for RDS for MySQL, RDS for PostgreSQL and Aurora MySQL and PostgreSQL. It recommends DB instance classes, storage type, IOPS and throughput, Graviton alternatives and Aurora Standard versus I/O-Optimized, and flags idle databases. It needs at least 30 hours of metrics and does not cover MariaDB, SQL Server or Oracle.
How do I get AWS RDS rightsizing recommendations?
Opt in to AWS Compute Optimizer and Cost Optimization Hub, ideally from the management account so every member account is covered. Recommendations appear after about 30 hours of CloudWatch metrics and refresh daily. Enable Performance Insights on important instances so memory pressure is considered, and turn on enhanced infrastructure metrics if you need a 93-day lookback.
Does changing an RDS instance class cause downtime?
Yes. AWS documents that an outage occurs while the DB instance class is modified. The change runs in the next maintenance window unless you apply it immediately, and a Multi-AZ deployment shortens the interruption to roughly a failover. Plan resizes like any production change and test the smaller class on a replica or a Blue/Green copy first.
Can you reduce RDS allocated storage?
Not in place. AWS states you can increase allocated storage but cannot deallocate it. Since November 2024, RDS Blue/Green Deployments can shrink storage for PostgreSQL 12 and later, MySQL 5.7 and later and MariaDB 10.4 and later, with the new size at least 20 percent above current usage, and the switchover takes about a minute.
Does moving RDS from gp2 to gp3 save money?
Not per gigabyte. On RDS in US East, gp2 and gp3 both cost $0.115 per GB-month on Single-AZ and $0.23 on Multi-AZ. The saving with gp3 comes from buying IOPS and throughput separately instead of over-allocating storage to get performance, so check whether you sized storage for IOPS before expecting a cut.
How much does RDS Extended Support cost?
$0.10 per vCPU-hour in years one and two after standard support ends and $0.20 per vCPU-hour in year three, in US East. Multi-AZ bills the standby too, which doubles it. A 4-vCPU db.r5.xlarge costs about $292 a month extra in years one and two and $584 in year three. MySQL 8.0 standard support ended on 31 July 2026.
Should I rightsize RDS before buying Reserved Instances or Database Savings Plans?
Yes. Database Savings Plans, launched in December 2025, are one-year hourly commitments of up to 20 percent off provisioned instances and up to 35 percent off serverless, and a later downsize does not lower the commitment. Reserved Instances are size flexible for MySQL, MariaDB, PostgreSQL, Oracle BYOL and Aurora, but not for SQL Server or Oracle license included.
When is Aurora I/O-Optimized cheaper than Aurora Standard?
When I/O is about 25 percent or more of your Aurora bill, according to AWS. I/O-Optimized removes the $0.20 per million request I/O charge but costs about 30 percent more for instances and $0.225 instead of $0.10 per GB-month for storage. Compute Optimizer now recommends the switch per cluster.
Is Graviton cheaper for RDS?
Usually by around 10 percent at list price for the same size. In US East a db.r6g.large for PostgreSQL costs $0.225 an hour against $0.25 for a db.r6i.large. AWS benchmarks Graviton4 instances at up to 29 percent better price performance than Graviton3, and Compute Optimizer shows the Graviton price when you set that architecture preference.
Keep reading
See your savings in dollars
Upload your cloud billing export and get a prioritized savings plan. Money never moves. Create your account and get started.