AWS Memory-Based Databases: Curriculum Overview

This curriculum covers the identification and application of AWS in-memory database services, specifically focusing on Amazon ElastiCache and Amazon MemoryDB for Redis. It is designed to prepare students for the AWS Certified Cloud Practitioner (CLF-C02) exam and practical cloud architecture tasks.

Prerequisites

Before starting this module, students should have a baseline understanding of the following:

• Cloud Computing Fundamentals: Understanding the difference between on-premises and cloud-based infrastructure.
• Basic Database Concepts: Familiarity with the difference between Relational (SQL) and Non-relational (NoSQL) databases.
• Storage Tiers: Awareness of the performance gap between disk-based storage (SSD/HDD) and Random Access Memory (RAM).
• Network Latency: Understanding how data retrieval times affect user experience in web applications.

Module Breakdown

Learning Objectives per Module

Module 1: Intro to In-Memory Computing

• Explain why in-memory databases provide sub-millisecond latency.
• Contrast disk-based retrieval (Amazon RDS) with memory-based retrieval (ElastiCache).

Module 2: Amazon ElastiCache

• Identify the two supported engines: Redis and Memcached.
• Determine when to use Memcached for simplicity vs. Redis for complex data types.

Module 3: Amazon MemoryDB for Redis

• Differentiate between ElastiCache (caching only) and MemoryDB (durable primary database).
• Describe how MemoryDB uses a multi-AZ transactional log for data persistence.

Module 4: Architectural Patterns

• Explain the Lazy Loading and Write-Through caching strategies.
• Identify the use of Time to Live (TTL) for cache expiration.

Visual Anchors

Caching Architecture Flow

Database Engine Comparison

Formula / Concept Box

[!IMPORTANT] The Performance Equation In-memory databases operate on the principle of reducing $I/O$ overhead: $Latency_{Total} = Latency_{Network} + Latency_{Processing} + Latency_{Storage}$ By moving storage to RAM, $Latency_{Storage} drops from milliseconds (10^{-3}$) to microseconds ($10^{-6}$).

Success Metrics

To demonstrate mastery of this curriculum, the student must be able to:

1. Engine Selection: Correctly identify whether an application requires the multi-threaded simplicity of Memcached or the advanced data structures (sets, lists) of Redis.
2. Service Identification: Differentiate between Amazon RDS (Relational), DynamoDB (NoSQL), and ElastiCache (In-memory) in a mock exam scenario.
3. Durability Analysis: Explain why ElastiCache is typically used as a "side-car" to a database while MemoryDB can stand alone.

Real-World Application

• Gaming Leaderboards: Using Redis "Sorted Sets" to rank millions of players in real-time with zero lag.
• E-commerce Session Stores: Keeping shopping cart data in ElastiCache so users don't lose items if a web server restarts.
• Database Offloading: Reducing the cost of expensive RDS instances by caching the results of frequent SQL queries (e.g., "Top 10 Products").

Examples Section

Case Study 1: The Social Media Feed

An application like Instagram uses in-memory caching to store the "top posts" of the hour. Instead of hitting the main database every time a user refreshes their feed, the application pulls from an ElastiCache (Redis) cluster.

• Result: Reduced database load by 80% and faster page loads for users.

Case Study 2: Financial Transaction Logging

A fintech startup needs a fast Redis-compatible interface but cannot afford to lose a single transaction if a node fails. They choose Amazon MemoryDB because it provides the speed of Redis with the durability of a relational database.

[!TIP] Remember for the exam: If the question mentions sub-millisecond latency or caching, the answer is almost always ElastiCache.