Overview

Amazon RDS is a managed relational database service that handles routine provisioning, patching, backups, and scaling across eight database engines. It suits teams already building on AWS that want managed operations without committing to a single engine. Its main advantage is combining engine choice with recovery and failover options; its main drawback is a bill shaped by separate usage charges and deployment choices.

Key features

Engine choice and managed operations

RDS supports Aurora PostgreSQL-Compatible, Aurora MySQL-Compatible, PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, and Db2. That range makes it useful for teams with different engine requirements or existing application dependencies. The service automates core database administration, but teams still need to select and pay for the capacity and deployment configuration their workloads require.

Recovery and availability

Automated backups are enabled by default, with point-in-time recovery up to the configured retention period, capped at 35 days. That provides a straightforward recovery path, though it is a bounded window rather than long-term archival. Multi-AZ deployments synchronously replicate data to standby instances and fail over automatically, with an availability SLA up to 99.95%. This is the stronger fit for workloads where continuity matters, but it is excluded from the Free Tier.

Read scaling and capacity

Read replicas make asynchronous copies for read-heavy workloads and can be promoted to standalone DB instances. Several engines support up to 15 replicas, making this a practical option when reads need to scale separately; it does not replace synchronous standby failover. Push-button compute scaling reaches up to 128 vCPUs, while some engines support storage up to 256 TiB. Those ceilings give growing workloads room to expand, but actual limits depend on engine and configuration.

Security, monitoring, and access

RDS supports VPC isolation, security groups, IAM permissions, SSL/TLS in transit, and AWS KMS encryption at rest using AES-256. All engines are HIPAA-eligible when used under an executed AWS Business Associate Agreement. CloudWatch, CloudWatch Logs, CloudTrail, Performance Insights, and Database Activity Streams support monitoring and audit workflows. Teams can manage instances through the AWS console, CLI, APIs, SDKs, or CloudFormation, and AWS RDS Ready validates partner products for areas such as monitoring, security, tuning, migration, and business applications.

Pricing

AWS Free Tier for RDS

0.00 USD per free. Billed Free Tier credits are valid for up to 12 months. New sign-ups receive $100 in credits plus up to $100 additional credits; legacy accounts may receive 750 hours per month for up to 12 months. The legacy offer includes select Single-AZ instances, 20 GB storage, and 20 GB automated backups. This is suitable for evaluating eligible configurations, not a reliable basis for budgeting a production deployment: custom engine options, Reserved Instances, non-Single-AZ deployments, snapshot migration, Query Editor, and Aurora read replicas are excluded.

On-Demand Instances

Custom pricing. Billing is per second while an instance runs, with no long-term commitment. Compute, storage, backup, and data transfer charges apply separately. This structure fits variable workloads or teams that value flexibility, but the total is not captured by an instance rate alone.

Reserved Instances

Custom pricing. Choose a 1-year or 3-year term with No Upfront, Partial Upfront, or All Upfront payment options. The discount applies to hourly instance usage; storage, backup, and I/O are billed separately. This can suit predictable usage when a commitment is acceptable, but it does not lock in the full database bill.

Platforms

RDS is available through API and web interfaces. It is a managed service, supports both deployment models, offers read replicas, and supports JSON data.

Who it's for

RDS is a strong fit for AWS teams that want to keep relational databases managed while choosing among widely used engines, or need built-in backup recovery, Multi-AZ failover, and read-replica options. It is less compelling for buyers who need a fixed all-in price, a generous free production configuration, or workloads that do not benefit from AWS's managed database model.

Pros and cons

Broad engine coverage: eight supported engines give teams room to match application needs without switching managed services.
Recovery and failover options: default backups, point-in-time recovery, and Multi-AZ automatic failover cover distinct continuity needs.
Separate read scaling: asynchronous replicas can serve read-heavy workloads and be promoted when needed.
Usage costs require attention: compute, storage, backup, data transfer, and—in Reserved Instances—IO charges are separate, so instance pricing alone understates spend.
Free Tier is constrained: several notable options, including Multi-AZ and Aurora read replicas, are excluded.

Alternatives

Choose MySQL if a freemium option and self-hosted platform support matter more than RDS's managed-service scope. For database monitoring rather than database hosting, Percona Monitoring and Management offers a free, open-source monitoring and observability product, though installation infrastructure may cost extra. Atatus API Analytics is a monitoring-focused alternative with custom pricing based on infrastructure size, data volume, and needs, plus a 14-day trial.

Google Cloud SQL is a paid alternative with a Cloud SQL Enterprise plan at 0.04 USD per month, up to 96 vCPUs, and a 99.95% SLA; its point-in-time recovery logs are retained up to 7 days. PostgreSQL teams evaluating monitoring can consider pganalyze, whose Production plan costs 149.00 USD per month. SolarWinds Database Performance Analyzer is another paid performance-analysis option, offered by subscription or volume licensing with a quote required. For self-hosted PostgreSQL monitoring, pgwatch is free open-source software supporting PostgreSQL 14 and later. Grafana Cloud Application Observability is a freemium observability alternative with an always-free tier including 2,232 host hours.

Browse Database Software or Relational Databases for more database options. Teams focused on operational tooling can compare Database Performance Monitoring Software, Database Proxy Software, Spatial Databases, and Service Virtualization Software.

Verdict

Choose Amazon RDS if your team is on AWS and wants broad engine choice with managed backup, failover, and read-scaling options. The operational coverage is the central reason to buy; look elsewhere if you need predictable all-in pricing or a free tier that includes higher-availability deployments and the full feature set.

	Amazon RDS	Matisse
Free plan	Yes	No
Free trial	Yes	No
Paid from	Free	—
Open source	No	No
Platforms	api, Web	Windows, macOS, Linux
Free plan	No	No
Deployment model	both	self_hosted
Managed service	Yes	—
High availability	Yes	Yes
Read replicas	Yes	—
JSON data support	Yes	—

Amazon RDS vs Matisse

Amazon RDS

Matisse