Skip to content
Explainer · AWS

What Is AWS RDS? Amazon RDS Explained (Engines, Pricing, Limits)

Amazon RDS (Amazon Relational Database Service) is AWS's managed service for running PostgreSQL, MySQL, MariaDB, SQL Server, Oracle and Db2. AWS provisions the server, installs and patches the engine, takes backups and handles failover; you choose the engine, instance size and storage and look after your schema and queries. This hub explains how RDS works and links to the detailed guides.

Facts checked 8 October 2026 against the providers' official documentation and pricing pages. Next review due January 2027. Cloud services, regions and prices change often; confirm on the provider's site before you buy.
Short answer
  • Amazon RDS runs six relational engines as a managed service: IBM Db2, MariaDB, Microsoft SQL Server, MySQL, Oracle Database and PostgreSQL. Aurora is a separate, related service.
  • AWS handles provisioning, patching, backups, failure detection and recovery; you remain responsible for query tuning and database design.
  • You pay per instance-hour plus storage, backup storage above a free allowance and data transfer; a small PostgreSQL instance lists at 0.016 USD per hour in US East (N. Virginia).
  • High availability comes from Multi-AZ deployments; read scaling from up to 15 read replicas; recovery from automated backups (up to 35 days) and snapshots.
  • Default quotas include 40 DB instances per Region and storage up to 64 TiB per instance; there is no operating system access except through RDS Custom.
How we know: Research-based: engines, features, limits and quotas were checked against the Amazon RDS product page, pricing page and FAQs and the Amazon RDS User Guide on 8 October 2026; the example prices come from the AWS price data behind the RDS pricing pages. We have not created RDS instances or taken console screenshots for this guide.

What is AWS RDS (Amazon Relational Database Service)?

AWS RDS is a managed relational database service. The Amazon RDS User Guide says RDS manages backups, software patching, automatic failure detection and recovery. The basic unit is the DB instance, an isolated database environment that can hold one or more databases, which you create and modify through the console, the AWS CLI or the RDS API and connect to with the same tools you would use for a standalone server.

The User Guide compares three models. On premises you manage everything. On Amazon EC2, AWS manages the hardware but you still install, patch, back up and fail over the database yourself. On RDS, AWS takes over scaling, high availability, backups, engine and operating system patching and installation, while you keep application optimisation. AWS recommends RDS as the default choice for most relational deployments, but the shared responsibility model is explicit that query tuning remains yours.

Who does what: self-managed on EC2 vs Amazon RDS (from the RDS User Guide)
TaskDatabase on Amazon EC2Amazon RDS
Application and query optimisationYouYou
Scaling, high availability, backupsYouAWS
Database software install and patchingYouAWS
Operating system install and patchingYouAWS
Server maintenance, hardware, power and networkAWSAWS

RDS database engines

RDS supports six engines. Features and versions vary by engine and Region, and each engine has its own pricing page and licensing model. Aurora (MySQL- and PostgreSQL-compatible, plus Aurora DSQL) is sold under the RDS family but uses a different storage architecture; see the Aurora guide in the related links.

Amazon RDS engines (October 2026)
EngineLicensing on RDSNotes
PostgreSQLOpen source, no licence feeSupports Multi-AZ DB clusters with two readable standbys
MySQLOpen source, no licence feeSupports Multi-AZ DB clusters with two readable standbys
MariaDBOpen source, no licence feeFree tier eligible, like PostgreSQL and MySQL
Microsoft SQL ServerLicence included (Express, Web, Standard, Enterprise); bring your own media is also availableExpress Edition is the only free tier edition
Oracle DatabaseLicence included or bring your own licenceUp to 15 read replicas, AWS recommends 5
IBM Db2Bring your own licence or licence through AWS MarketplaceNot free tier eligible

Instance classes and storage

A DB instance class sets CPU and memory. RDS groups classes into general purpose (db.m*), memory optimised (db.r*, db.x*, db.z*), compute optimised (db.c*) and burstable (db.t*), with AWS Graviton processors indicated by the "g" in names such as db.m7g and db.m8g. You can change the class later by modifying the instance.

Storage is Amazon EBS: General Purpose SSD (gp3 and gp2) for most workloads, or Provisioned IOPS SSD (io1, io2) for I/O-intensive production databases. The pricing page lists General Purpose capacity from 20 GiB to 64 TiB. Magnetic storage is deprecated: it is no longer offered for new instances, and from 1 July 2026 snapshots can no longer be restored to it.

High availability, read replicas and backups

Multi-AZ. A Multi-AZ DB instance deployment keeps a synchronous standby in a second Availability Zone that RDS fails over to automatically; a Multi-AZ DB cluster (PostgreSQL and MySQL) adds two readable standbys in three AZs. Read replicas are asynchronous, read-only copies for scaling reads or disaster recovery, including in other Regions. The differences and costs are in RDS Multi-AZ vs read replicas.

Backups. Automated backups allow point-in-time recovery within a retention period of up to 35 days, and manual snapshots are kept until you delete them. Every restore creates a new instance; the steps are in RDS backups and snapshots.

Connections. For applications that open many short-lived connections, such as Lambda functions, RDS Proxy pools connections and shortens failover.

AWS RDS pricing and free tier in brief

RDS has no minimum fee. You pay for instance hours (billed per second after a 10-minute minimum), provisioned storage, backup storage above the free allowance, data transfer and optional features. Multi-AZ roughly doubles instance and storage cost. Reserved Instances and Database Savings Plans reduce the instance rate in return for a commitment. Full details and three worked examples are in AWS RDS pricing.

New AWS accounts now get credits rather than 12 months of free instance hours: 100 USD on sign-up and up to 100 USD more, usable on RDS micro instances among other services. Older accounts keep the legacy 750-hour offer until their 12 months end. See the RDS free tier guide.

RDS limits and quotas

Quotas apply per account and per Region. Most can be raised through Service Quotas; the table lists the defaults documented in the RDS User Guide.

Selected Amazon RDS default quotas and limits
LimitDefaultNotes
DB instances per Region40Shared with Aurora, Neptune and DocumentDB instances; adjustable
Read replicas per primary15Adjustable for RDS
Manual DB snapshots per Region100Automated backups do not count
Total storage for all DB instances per Region100,000 GBAdjustable
Storage per instance (General Purpose or Provisioned IOPS)Up to 64 TiBMinimum 20 GiB for General Purpose
Backup retention0 to 35 daysManual snapshots never expire
Reserved DB instances per Region40Adjustable
RDS Proxy proxies per Region20Adjustable
Stopping an instanceUp to 7 consecutive daysRestarts automatically afterwards

When to choose RDS, and when not to

RDS suits teams that want a standard PostgreSQL, MySQL, MariaDB, SQL Server, Oracle or Db2 database without running servers, with predictable instance-based pricing and AWS-managed backups and failover.

Look elsewhere when:

Pricing

Prices from the provider's official pricing pages, checked 8 October 2026, region US East (N. Virginia), in USD, excluding tax. List prices only; discounts, commitments and your actual usage change the bill.

Headline RDS for PostgreSQL On-Demand rates: db.t4g.micro 0.016 USD per hour and db.m8g.large 0.168 USD per hour (Single-AZ), and gp3 storage 0.115 USD per GB-month. That puts a micro instance with 20 GB at about 13.98 USD a month (730 hours x 0.016 + 20 x 0.115) before tax and data transfer. Other engines, licence-included editions and Regions differ; see AWS RDS pricing for the breakdown.

Frequently asked questions

What is RDS in AWS?

Amazon RDS (Relational Database Service) is AWS's managed service for relational databases. It runs PostgreSQL, MySQL, MariaDB, SQL Server, Oracle and Db2 on instances that AWS provisions, patches, backs up and fails over for you.

Is Amazon RDS the same as Aurora?

No. Aurora is part of the RDS family and uses the same console and APIs, but it is a separate engine family (MySQL- and PostgreSQL-compatible) with its own distributed storage and pricing. The RDS User Guide covers engines other than Aurora.

Is AWS RDS a database or a service?

A service. RDS runs existing database engines for you; the database itself is PostgreSQL, MySQL or one of the other supported engines, so standard drivers and tools connect to it.

Can I access the server behind an RDS instance?

No, RDS does not give operating system access. If an application needs OS-level customisation, AWS offers RDS Custom (for SQL Server; RDS Custom for Oracle reaches end of support on 31 March 2027) or you can run the database on EC2.

How much does Amazon RDS cost?

In US East (N. Virginia) a db.t4g.micro PostgreSQL instance lists at 0.016 USD per hour and gp3 storage at 0.115 USD per GB-month (October 2026), so a small database starts at roughly 14 USD a month. Production instances with Multi-AZ typically cost hundreds of dollars a month.

What is the maximum storage for an RDS instance?

64 TiB for General Purpose SSD and Provisioned IOPS SSD storage, according to the RDS pricing page. Aurora has a separate limit.

Sources

Checked 8 October 2026.

How we research cloud database guides: our editorial method. CodeWithSQL earns nothing from the providers mentioned.

Choosing where to run your database?

Start with the section overview, or compare providers side by side.