AWS Databases & Analytics: A Complete Guide
AWS database services appear frequently on the CLF-C02 exam in the form of "which service should you choose for this situation?" scenarios. Understanding the key characteristics and use cases for each service is essential for exam success.
---
Relational DB vs NoSQL — Core Concepts
Relational Databases (SQL)
Data stored in a fixed schema (table structure) Data queried and managed via SQL Use cases: financial systems, transaction processing Examples: MySQL, PostgreSQL, Oracle, SQL Server
NoSQL Databases
Flexible schema (no predefined structure required) Optimized for horizontal scaling (Scale-Out) Use cases: real-time apps, IoT, mobile Examples: DynamoDB, DocumentDB
SQL is suited for structured data and transactional processing, while NoSQL is the right choice when you need a flexible schema or massive horizontal scaling.
---
Amazon RDS — The Standard for Relational Databases
RDS (Relational Database Service) is a fully managed relational database service operated and maintained by AWS. AWS handles patching, backups, and high-availability configuration automatically, allowing developers to focus on application logic.
Supported Engines
MySQL, PostgreSQL, MariaDB, Oracle, SQL Server, Amazon Aurora
EC2 Direct Installation vs RDS
| Item | EC2 Direct | RDS | |------|------------|-----| | OS Patching | Manual | AWS automatic | | Backups | Manual setup | Automatic + point-in-time recovery | | High Availability | Manual setup | Multi-AZ built-in | | Read Replica | Manual setup | Up to 5, easy to create | | SSH Access | Available | Not available |
RDS does not allow SSH access because AWS directly manages the underlying infrastructure. This point is frequently tested.
RDS Deployment Options
| Option | Purpose | Key Features | |--------|---------|-------------| | Read Replicas | Read performance | Async replication, up to 5, writes go to primary only | | Multi-AZ | High availability | Sync replication, automatic failover, one standby | | Multi-Region | Disaster recovery + global reads | Protects against region-wide failure, extra replication cost |
These three options serve distinct purposes. Read Replicas improve read performance, Multi-AZ guards against Availability Zone failures, and Multi-Region provides disaster recovery against a full region outage.
---
Amazon Aurora — The Premium Edition of RDS
Aurora is a high-performance relational database developed by AWS. It is compatible with MySQL and PostgreSQL while delivering significantly higher performance and availability.
| Feature | Details | |---------|--------| | Compatibility | MySQL and PostgreSQL compatible | | Performance | 5x faster than MySQL, 3x faster than PostgreSQL | | Storage | Auto-scaling (up to 64TB) | | Read Replicas | Up to 15 supported | | Cost | 20% more than RDS, but more efficient |
Aurora is the right choice for enterprise-grade workloads that demand high performance and high availability.
---
Amazon ElastiCache — In-Memory Caching
ElastiCache caches frequently used query results in memory, reducing database load and improving response times. AWS handles OS patching and backups automatically as a fully managed service.
| Engine | Features | |--------|----------| | Redis | Replication and persistence support, advanced features | | Memcached | Simple memory cache, fast and lightweight |
ElastiCache sits in front of your database as a high-performance in-memory cache, dramatically reducing response times for repeated read requests.
---
Amazon DynamoDB — Serverless NoSQL
DynamoDB is AWS's fully managed serverless NoSQL database. It can handle millions of requests without any infrastructure management, and supports trillions of rows and hundreds of terabytes of data.
| Feature | Details | |---------|--------| | Data Model | Key-value + document | | Latency | Single-digit milliseconds | | Scalability | Millions of req/sec, trillions of rows, hundreds of TB | | Availability | Automatically replicated across 3 AZs | | Management | Serverless — no infrastructure management needed |
DynamoDB Accelerator (DAX)
A purpose-built in-memory cache for DynamoDB that reduces response times from milliseconds to microseconds (μs).
DynamoDB Global Tables
Multi-region replication (multi-master) Low-latency reads and writes from anywhere in the world
DynamoDB's defining characteristics are serverless + NoSQL + single-digit millisecond latency. Adding DAX brings responses down to the microsecond level.
---
Analytics and Data Warehouse Services
Amazon Redshift — Data Warehouse
OLAP (Online Analytical Processing) — optimized for large-scale analytical queries Columnar storage → fast aggregation queries 10x faster performance than other data warehouses, petabyte scale SQL interface, integrates with QuickSight and Tableau
Use RDS for transactional processing (OLTP) and Redshift for large-scale analytics (OLAP).
Amazon Athena — Serverless SQL Queries
Directly query data stored in S3 using serverless SQL No infrastructure to manage, pay only for data scanned ($5/TB) Supports CSV, JSON, Parquet, ORC and other formats Use cases: VPC Flow Logs, ELB logs, CloudTrail analysis
Whenever a scenario asks about analyzing S3 data with SQL, Athena is the answer.
Amazon EMR — Big Data Processing
Managed big data clusters based on Hadoop, Spark, and Hive Automatically provisions clusters of hundreds of EC2 instances Use cases: large-scale ETL, machine learning, web indexing
Amazon QuickSight — BI Dashboard
Serverless business intelligence (BI) visualization tool Connects to RDS, Redshift, S3, and other data sources Per-session pricing model, ML-powered insights Use cases: executive dashboards, data visualization
AWS Glue — Serverless ETL
Fully managed ETL (Extract, Transform, Load) service Serverless, prepares and transforms data for analytics Glue Data Catalog: a data catalog that integrates with Athena, Redshift, and EMR
---
Purpose-Built Databases
DocumentDB — MongoDB-Compatible Document DB
AWS-developed MongoDB-compatible managed database, similar to Aurora Optimized for JSON document storage, Multi-AZ, auto-scaling storage (up to 64TB) Use cases: content management, catalogs, mobile backends
Amazon Neptune — Graph Database
Fully managed graph database Queries billions of relationships with millisecond latency Use cases: social networks, fraud detection, recommendation engines, knowledge graphs (Wikipedia)
Amazon QLDB — Ledger Database