AWS Exam Guide: EC2 Self-Hosted vs. Amazon RDS vs. Amazon Aurora (MySQL)

1. Summary Comparison Table

Feature / Dimension	Self-Hosted MySQL on EC2	Amazon RDS MySQL	Amazon Aurora MySQL
Service Category	IaaS (Infrastructure as a Service)	PaaS (Platform as a Service)	Cloud-Native PaaS
OS / Root Access	YES (Full root access to host OS)	NO (AWS manages underlying OS)	NO (AWS manages underlying OS)
Management Overhead	Highest (Customer manages everything)	Low (AWS manages OS/patching/backups)	Lowest (AWS manages OS/patching/storage scaling)
Performance	Standard (depends on EC2/EBS selection)	Standard (native MySQL performance)	High Performance (Up to 5x standard MySQL)
Storage Architecture	Manual EBS volume management	Pre-provisioned EBS storage (auto-expand available)	Auto-scaling storage (grows in 10 GB increments up to 128 TiB)
High Availability (HA)	Manual setup (EC2 replication, scripts)	Multi-AZ synchronous replication (Standby in 2nd AZ)	Automatic 6 copies across 3 AZs (Fault-tolerant, self-healing)
Read Replicas	Manual setup	Up to 5 Read Replicas	Up to 15 Read Replicas (ultra-fast, low replication lag)

Understanding who handles what is a primary focus for the CLF-C02 exam:

AWS manages: Physical hardware, host virtualization, facility security.
Customer manages: Guest OS installation & patching, database software installation & tuning, database patching, automated backups, disaster recovery, and Multi-AZ replication setup.

AWS manages: OS installation & security patching, database software provisioning & patching, hardware maintenance, automated backups, point-in-time recovery, and underlying storage replication.
Customer manages: Database schema design, table creation, user authentication/permissions, query optimization, and network access rules (Security Groups).

Keywords: "Full control over OS", "root access required", "unsupported database version", "custom database extensions/plugins".
Exam Rule: If a question mentions needing OS root access or installing custom third-party kernel modules alongside the database, choose EC2.

Keywords: "Managed relational database", "offload OS maintenance and backups", "standard open-source MySQL engine", "Multi-AZ for high availability".
Exam Rule: Choose RDS when a standard relational database is required without OS administration overhead, but cloud-native enterprise performance is not specifically demanded.

Keywords: "Up to 5x performance of MySQL", "cloud-native relational database", "auto-scaling storage up to 128 TiB", "6 copies of data across 3 AZs", "serverless relational DB".
Exam Rule: Choose Aurora when maximum performance, high availability, fast failover, or auto-scaling relational storage is required.

Trap 1: OS Root Access
Neither Amazon RDS nor Amazon Aurora provides SSH or root access to the underlying OS. If a question states that the database administrator must have OS access, the answer is always Self-Hosted on EC2.
Trap 2: Performance Multipliers
AWS specifically quantifies Aurora performance: Up to 5x faster than standard MySQL and up to 3x faster than standard PostgreSQL.
Trap 3: Storage Scaling
Standard RDS requires choosing or configuring EBS volumes. Aurora automatically scales storage in increments as data grows, requiring no storage pre-provisioning.
Trap 4: Multi-AZ vs. Read Replicas
Multi-AZ: Used for High Availability (HA) and Disaster Recovery (DR). Writes go to primary; standby is synchronous.
Read Replicas: Used for Read Scaling (offloading read-heavy workloads from the primary DB). Writes do not go directly to read replicas.