Lesson 35

Relational Databases

Amazon RDS and Amazon Aurora

1. Managed vs. Unmanaged Databases

Running a standalone relational database on-premises or on EC2 requires you to manage server maintenance, OS patches, database software installation, backups, high availability, and scaling. Amazon RDS offloads these tasks to AWS so you can focus on application optimization.

Responsibility	On-Premises	EC2-Hosted	Amazon RDS
Application optimization	You	You	You
Scaling / High Availability	You	You	AWS
Database backups	You	You	AWS
Database software patches	You	You	AWS
OS installation and patches	You	You	AWS
Server maintenance / Power / Rack	You	AWS	AWS

2. Amazon RDS

Amazon Relational Database Service is a managed service that sets up, operates, and scales a relational database in the cloud. It provides cost-efficient, resizable capacity while automating time-consuming administration tasks.

Core idea: RDS lets you run relational databases without managing the underlying infrastructure. AWS handles backups, patching, high availability, and scaling.

Supported Database Engines

MySQL
MariaDB
PostgreSQL
Oracle
Microsoft SQL Server
IBM DB2

RDS DB Instances

The basic building block of RDS is the database instance — an isolated database environment that can contain multiple user-created databases. Each instance has a DB instance class (CPU, memory, network) and storage type (Magnetic, General Purpose SSD, or Provisioned IOPS). Instances typically run in a private VPC subnet accessible only to application instances.

Multi-AZ Deployment

After enabling Multi-AZ, RDS automatically creates a standby copy of the database instance in another Availability Zone within the same VPC. Transactions are synchronously replicated to the standby.

Exam tip: If the primary instance fails, RDS automatically fails over to the standby. Your application references the database by a DNS endpoint, so no code changes are needed.

Read Replicas

Read replicas use asynchronous replication from the source database. They offload read queries from your primary instance and support read-heavy workloads. Read replicas can be created in different Regions and can be promoted to the primary manually.

Feature	Multi-AZ	Read Replica
Replication type	Synchronous	Asynchronous
Primary purpose	High availability / DR	Read scaling
Failover	Automatic	Manual promotion

When to Use RDS

Complex transactions or complex queries
Medium to high query/write rate (up to 30,000 IOPS)
No more than a single worker node or shard
High durability requirements

When NOT to use RDS: Massive read/write rates (>150,000 writes/sec), sharding needs, simple GET/PUT queries (use DynamoDB), or RDBMS customization (use EC2-hosted).

3. Amazon Aurora

Amazon Aurora is a MySQL- and PostgreSQL-compatible relational database built for the cloud. It combines the performance and availability of high-end commercial databases with the simplicity and cost-effectiveness of open-source databases.

Key Features

High availability: Stores multiple copies of data across multiple Availability Zones with continuous backups to S3. Supports up to 15 read replicas.
Instant crash recovery: Restarts in less than 60 seconds in most cases. Buffer cache survives restart.
Fully managed: Automates provisioning, patching, backup, recovery, failure detection, and repair.
Compatible: Drop-in compatibility with MySQL and PostgreSQL. Use existing database tools with little or no change.
Security: Network isolation via VPC, encryption at rest via AWS KMS, encryption in transit via SSL.

Exam tip: Aurora is a managed service under Amazon RDS. It is designed for high performance, high availability (6 copies across 3 AZs), and compatibility with MySQL/PostgreSQL.

4. Quick Quiz

Primary Source: AWS Academy Module 8: Databases (module-8.pdf).

Ask your teacher: If you are unclear on the difference between RDS Read Replicas and Multi-AZ, or when to use RDS versus Aurora, ask before continuing. The exam frequently tests these distinctions.