Domain 3: Database Services

CLF-C02 Exam Domain 3 - Part 3 | 34% of Scored Content

Learning Objectives

By the end of this section, you will be able to:

• Understand Amazon RDS features and capabilities
• Compare RDS, DynamoDB, and ElastiCache
• Identify use cases for different database services
• Understand Multi-AZ and Read Replicas
• Compare SQL and NoSQL databases

Amazon RDS (Relational Database Service)

Overview

Amazon RDS is a managed relational database service that provides cost-efficient and resizable capacity.

Key Characteristics

Feature	Description
Managed Service	AWS handles provisioning, patching, backup, recovery
Supports 6 Engines	MySQL, PostgreSQL, Oracle, SQL Server, MariaDB, Aurora
Scalable	Scale compute and storage independently
High Availability	Multi-AZ deployments
Automatic Backups	Point-in-time recovery

RDS Components

1. DB Instances

Definition: An isolated database environment in the cloud.

Instance Classes:

• Burstable: t3.micro, t3.small (development, testing)
• General Purpose: m5 (balanced)
• Memory Optimized: r5, x1e (high memory requirements)
• Compute Optimized: c5 (high CPU)

2. DB Engine Options

Engine	Type	Use Case	License Included
MySQL	Open source	Web applications	Yes
PostgreSQL	Open source	Enterprise apps, GIS	Yes
MariaDB	Open source	MySQL-compatible	Yes
Oracle	Commercial	Enterprise applications	No (License Required)
SQL Server	Commercial	Windows applications	No (License Required)
Aurora	Proprietary	High-performance, cloud-native	Yes

RDS High Availability Features

1. Multi-AZ Deployment

Purpose: High availability and durability.

How It Works:

Primary DB Instance (AZ1)
         │
         └─ Synchronous Replication
                    │
         ┌──────────┴──────────┐
         ▼                     ▼
   Standby Instance (AZ2)   Automatic Failover

Characteristics:

• Synchronous replication to standby instance
• Automatic failover to standby if primary fails
• Same endpoint for applications (no DNS change)
• Separate AZ for standby

Use Cases: Production databases, critical applications

Note: Multi-AZ is for HA, not read scaling

2. Read Replicas

Purpose: Scale read capacity and improve performance.

How It Works:

Primary DB Instance
    │ (Write)
    ├─ Read Replica 1 (AZ1) ──┐
    ├─ Read Replica 2 (AZ2) ──┤─ Read Traffic
    └─ Read Replica 3 (AZ3) ──┘

Characteristics:

• Asynchronous replication from primary
• Up to 15 read replicas (MySQL, PostgreSQL)
• Separate endpoints for each replica
• Can be in different regions
• Must be in same region as Multi-AZ (if enabled)

Use Cases:

• Read-intensive applications
• Reporting and analytics
• Disaster recovery (cross-region replicas)

Note: Read Replicas are for read scaling, not HA

3. Automated Backups

Features:

• Point-in-time recovery (PITR)
• Retention: 1-35 days
• Backup window: Configurable time
• Snapshots: Stored in S3

4. Database Snapshots

Types:

Type	Description	Use Case
Automated	During backup window	Point-in-time recovery
Manual	On-demand	Before changes, long-term backup

Characteristics:

• Stored in S3
• Can create new DB instance from snapshot
• Cross-region snapshot copy

RDS Security

Feature:

Security Feature	Description
VPC	Deploy within VPC for network isolation
Security Groups	Control access at network level
Encryption at Rest	Using AWS KMS
Encryption in Transit	SSL/TLS
IAM	Control who can manage RDS resources
Master User	Database authentication (username/password)

---

Amazon Aurora

Overview

Amazon Aurora is a fully managed relational database engine compatible with MySQL and PostgreSQL.

Key Characteristics

Feature	Aurora vs RDS
Performance	Up to 5x MySQL, 3x PostgreSQL
Storage	Auto-scaling up to 128 TB
Availability	2 replicas in 3 AZs (storage)
Replication	Faster (under 1 second)
Cost	Higher than RDS MySQL/PostgreSQL

Aurora Cluster Architecture

┌─────────────────────────────────────────────┐
│            Aurora Cluster                   │
├─────────────────────────────────────────────┤
│                                             │
│   Primary Instance (Writer)                 │
│   ┌─┬─┬─┬─┬─┬─┬─┬─┬─┐                    │
│   └─┴─┴─┴─┴─┴─┴─┴─┴─┘ (Shared Storage)     │
│       │       │       │                    │
│   Reader  Reader  Reader (Aurora Replicas) │
│                                             │
└─────────────────────────────────────────────┘

Features:

• Storage is distributed across 3 AZs (automatically)
• Up to 15 Aurora Replicas for read scaling
• Instant failover (typically < 30 seconds)
• Backtrack: Rewind database to specific time (up to 72 hours)

Aurora Serverless

Characteristics:

• On-demand database
• Auto-scaling compute capacity
• Pay per second
• Use Cases: Infrequent applications, unpredictable workloads

---

Amazon DynamoDB

Overview

Amazon DynamoDB is a fully managed NoSQL database service that provides fast and predictable performance.

Key Characteristics

Feature	Description
NoSQL	Key-value and document data
Managed	No server provisioning
Scalable	Unlimited throughput and storage
Fast	Single-digit millisecond latency
Pay-per-use	Pay for read/write capacity

DynamoDB Core Concepts

1. Tables

Definition: Collection of items.

Primary Keys:

• Partition Key (Hash Key): Unique value for each item
• Partition Key + Sort Key (Composite): Unique combination

2. Items

Definition: Individual data records (like rows in SQL).

Attributes: Key-value pairs (like columns in SQL)

Example Item:

{
  "UserId": "user123",           // Partition Key
  "Timestamp": "2024-01-15",     // Sort Key
  "Message": "Hello World",
  "Likes": 42
}

DynamoDB Capacity Modes

Mode	Description	Pricing
On-Demand	No capacity planning	Pay per request
Provisioned	Specify RCUs and WCUs	Pay for capacity

On-Demand:

• No capacity planning needed
• Handles unpredictable traffic
• Higher cost than provisioned

Provisioned:

• RCU: Read Capacity Units (1 RCU = 1 strongly consistent read per second for 4 KB)
• WCU: Write Capacity Units (1 WCU = 1 write per second for 1 KB)
• Auto Scaling: Adjusts based on traffic

DynamoDB Features

DAX (DynamoDB Accelerator):

• In-memory cache
• Up to 10x performance improvement
• Reduces read cost

TTL (Time To Live):

• Automatic data deletion
• Use Case: Session data, logs, time-series data

Streams:

• Capture item modifications
• Use Case: Triggers, replication, analytics

Global Tables:

• Multi-region, multi-master replication
• Use Case: Global applications, low latency

On-Demand Backup:

• Full backups anytime
• No performance impact
• Retained until deleted

---

Amazon ElastiCache

Overview

Amazon ElastiCache is a managed in-memory cache service.

Supported Engines

Engine	Type	Use Case
Redis	In-memory data store	Caching, pub/sub, leaderboards
Memcached	Simple memory cache	Simple caching, session state

ElastiCache Use Cases

Use Case	Description
Caching	Reduce database load
Session State	Store user sessions
Leaderboards	Game leaderboards (Redis sorted sets)
Pub/Sub	Real-time messaging (Redis)
Real-time Analytics	Counters, analytics

ElastiCache vs DynamoDB

Feature	ElastiCache	DynamoDB
Type	In-memory cache	Persistent database
Latency	Sub-millisecond	Single-digit milliseconds
Persistence	Optional (Redis)	Yes
Use Case	Cache, session state	Primary database

---

Database Service Comparison

SQL vs NoSQL

Aspect	SQL (RDS)	NoSQL (DynamoDB)
Schema	Fixed schema	Flexible schema
Scaling	Vertical	Horizontal
Query Language	SQL	Native API
Transactions	ACID	ACID (for single item)
Use Case	Structured data, complex queries	Rapid growth, flexible schema

When to Use Which

Use RDS when:

• You need relational database features
• Complex queries and joins
• ACID transactions across multiple tables
• Existing SQL applications

Use DynamoDB when:

• You need single-digit millisecond latency
• Your data access patterns are simple
• You need to scale throughput
• Your data model is simple (key-value)

Use ElastiCache when:

• You need to cache frequently accessed data
• You need session storage
• You need pub/sub functionality

---

Exam Tips - Database Services

High-Yield Topics

1. RDS Multi-AZ vs Read Replicas:

   • Multi-AZ = Synchronous replication, HA, same endpoint
   • Read Replicas = Asynchronous replication, read scaling, separate endpoints

2. Aurora:

   • AWS proprietary database
   • 5x faster than MySQL
   • Storage auto-scaling (up to 128 TB)
   • 2 replicas in 3 AZs (storage level)

3. DynamoDB:

   • NoSQL, key-value/document
   • Single-digit millisecond latency
   • On-demand vs Provisioned capacity
   • Partition Key (+ Sort Key)

4. ElastiCache:

   • In-memory cache
   • Redis and Memcached
   • Caching, session state, leaderboards

5. Backup/Restore:

   • RDS: Automated backups (1-35 days), snapshots
   • DynamoDB: On-demand backup, PITR (up to 35 days)

Additional Resources

DigitalCloud Training Cheat Sheets

• AWS Database Services Cheat Sheet - Comprehensive database services reference for exam prep

Official AWS Documentation

• Amazon RDS Documentation
• Amazon Aurora Documentation
• Amazon DynamoDB Documentation
• Amazon ElastiCache Documentation
• AWS Skill Builder - Free database courses and certification prep
• AWS Builder Labs - Hands-on database labs and practice environments
• AWS Database Learning Plans - Comprehensive database learning paths

AWS Database Resources

• Amazon RDS Pricing - Current database pricing
• Aurora Pricing - Cloud-native database pricing
• DynamoDB Pricing - NoSQL database pricing
• ElastiCache Pricing - In-memory cache pricing

---

Next: Networking Services