Databases
A managed database takes installation, patching, backups and replication off your hands. A database on EC2 gives full control, but all of that stays with you.
RDS
Relational Database Service
A managed relational database with automated backups, patching and high availability.
Key points
- Engines: Amazon Aurora, PostgreSQL, MySQL, MariaDB, Oracle and SQL Server.
- Multi-AZ: a standby replica in another AZ, with automatic failover.
- Read replicas to scale queries.
On the exam
A relational database without managing servers: RDS. Need access to the database operating system? Then it is EC2.
Aurora
Amazon Aurora
An AWS relational database compatible with MySQL and PostgreSQL, built for the cloud.
Key points
- Up to 5 times faster than standard MySQL and 3 times faster than standard PostgreSQL.
- Six copies of the data across three Availability Zones.
- Distributed, fault-tolerant, self-healing storage.
On the exam
A high-performance, highly available relational database compatible with MySQL or PostgreSQL: Aurora.
DynamoDB
Amazon DynamoDB
A serverless, fully managed NoSQL key-value and document database.
Key points
- Single-digit millisecond latency at any scale.
- Scales automatically, with no servers to manage.
- Global tables replicate data across Regions.
On the exam
NoSQL, serverless, milliseconds at any scale: DynamoDB.
ElastiCache
Amazon ElastiCache
A managed in-memory cache compatible with Valkey, Redis OSS and Memcached.
Key points
- Microsecond latency.
- Offloads the main database by keeping frequent queries in memory.
- Uses: user sessions, leaderboards, page caching.
On the exam
An in-memory database to speed up repeated reads: ElastiCache.
DocumentDB
Amazon DocumentDB
A managed JSON document database compatible with MongoDB.
Key points
- For catalogs, user profiles and content management.
- Storage replicated across three AZs.
- Eases moving MongoDB workloads without managing servers.
On the exam
"Move a MongoDB database to a managed service": DocumentDB.
Neptune
Amazon Neptune
A managed graph database for highly connected data.
Key points
- Relationship queries with Gremlin, openCypher and SPARQL.
- Uses: social networks, recommendations, fraud detection, knowledge graphs.
- Replication across three AZs.
On the exam
Relationships between entities (social networks, recommendations): Neptune.
Redshift
Amazon Redshift
A columnar data warehouse with massively parallel processing, for analytics at large volume.
Key points
- Analytical SQL queries over petabytes of data.
- Columnar storage and parallel processing.
- Integrates with data lakes on S3.
On the exam
Data warehousing and business intelligence over large volumes: Redshift. Day-to-day transactions belong in RDS or Aurora.
DMS
Database Migration Service
Migrates databases to AWS while the source database stays operational throughout.
Key points
- Homogeneous (Oracle to Oracle) and heterogeneous (Oracle to Aurora) migrations.
- The AWS SCT (Schema Conversion Tool) converts the schema between different engines.
- Continuous replication to minimize downtime.
On the exam
Migrate a database with minimal downtime: DMS. Changing engines: DMS + SCT.