Guide to Cost Factors for Amazon’s RDS Pricing

What is the Relational Database Service?

Amazon Relational Database Service (RDS) provides database engines of multiple types and instances. Users can manage and scale these databases to suit their needs and configure them with the most needed parts. Though Amazon does claim that its performance and scalability can match its competitors while remaining a mere fraction of the cost, this is the ideal circumstance with proper price control enacted. This guide will review all of the related price variables tied to RDS.

Just for reference, RDS is one of the services that can take advantage of AWS’ Free Tier category. For the first year after signing up with AWS, users receive 750 hours of database usage per month for MySQL, PostgreSQL, MariaDB, or SQL Servers using single-AZ db.t2.micro, db.t3.micro, and db.t4g.micro Instances. Additionally, users can access 20GB of SSD cloud storage and 20GB of storage dedicated to snapshots and backups. Regardless of how much they used this service during the trial period or if they even used it, this offering will expire after the first twelve months of the account’s lifespan, so use this offer to learn and adapt to the RDS pricing model.

As an advanced AWS partner, we bring unparalleled expertise to architect, deploy, and optimize cloud solutions tailored to your unique needs. 

Database Engine Pricing

Currently, RDS supports seven relational database engines with an eighth option for on-premises installation. Pricing does vary further with regard to the instance type:

• On-Demand Instances: Users have the flexibility to pay by the hour without any upfront fees or long-term commitments. If a database is utilized for less than an hour, billing is done in one-second increments with a minimum charge of 10 minutes. The billing starts the moment the instance is started and stops as soon as it is stopped. This option is the default for all engines.
• Reserved Instances: Amazon RDS Reserved Instances offer users the opportunity to secure significant discounts by committing to 1 or 3-year terms. The pricing structure accommodates users’ resource needs, ensuring flexibility and cost savings. When considering payment options, users can choose from three distinct plans to maximize their savings potential:
• • Opting for no upfront fees guarantees a 29% discount compared to the On-Demand rate.
  • Selecting a partial upfront payment, ranging from 0% to 99%, unlocks savings of up to 33% for 1-year plans and 52% for 3-year plans.
  • For those looking to make a full upfront payment, substantial savings of up to 34% on 1-year plans and 53% on 3-year plans are available. These options empower users to tailor their payment strategy to achieve the most advantageous cost benefits for their Amazon RDS Reserved Instances.

Amazon Aurora

MariaDB, MySQL, and PostgreSQL

Oracle and Microsoft SQL Server

Database Instance Pricing

Combined with the engine of choice, users will need to select an instance type, a vCPU for compute power, GiB RAM for memory, and networking options. Instance options range from db.t3.micro (has 2 vCPUS, 1 GiB RAM, and supports 2085 Mbps) to db.m5.24xlarge ( with 96 vCPUS, 384 GiB RAM, and does 19,000 Mbps). This process will involve extensive planning and considering what users need from the database.

Regional Pricing

Amazon distributes its services globally through Availability Zones (AZ). Each zone does have differing rates and outage protection options. Users will need to choose a region as well as consider the instance type, as the aforementioned rates are all dependent on the AZ of choice. Users do have the choice to deploy across multiple AZs simultaneously. These networks will consist of a primary DB with secondary DBs located at other AZs. This has the benefit of both providing low latency at fringe locations and data redundancy in the case of outages, failures, and traffic congestion.

Amazon RDS pricing varies based on the choice of database engine. Currently, RDS supports seven relational database engines with an eighth option for on-premises installation. Pricing does vary further on the instance type. On-Demand Instances allow users to pay by the hour, while Reserved Instances offer significant discounts for commitments of up to 1 or 3 years. The pricing model for each engine is different, with options for additional savings based on the user’s initial payment.

Amazon Aurora, the primary database engine, charges based on GB/month and I/O usage per million requests. MariaDB, MySQL, and PostgreSQL operate under the same pricing model as Amazon Aurora, with PostgreSQL On-Demand instances being up to 10% more expensive than Aurora depending on size. Oracle and Microsoft SQL Server choices are nearly double the cost of open-source options due to licensing, but users can bring their own Oracle license to align hourly rates with open-source options.

Storage Pricing

General Purpose SSD

Provisioned IOPS SSD

Magnetic Storage

Other Price Factors

• RDS Backup: Backups start at $0.095/GB per month, even after instance termination.
• Snapshot export: Exports cost $0.01/GB of snapshot size. This will export snapshots in the Parquet format to an Amazon S3 bucket. This format doubles export speeds and requires a sixth of the space.
• RDS data transfer: These rates will only apply to outbound transfers. Pricing is separated into tiers. It charges $0.09/GB for the first 10 TB/month and outbound transfers are free for the first 100 GB/month across all AWS regions and AZs outside of GovCloud and China.