DynamoDB

Learning outcomes

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

1. Explain what Amazon DynamoDB is.
2. Distinguish DynamoDB from a relational database like RDS.
3. Explain the basic DynamoDB data model: tables, items, and attributes.
4. Recognize the importance of the primary key in DynamoDB design.
5. Explain the difference between on-demand and provisioned capacity.

---

What DynamoDB is

DynamoDB is AWS’s serverless NoSQL database for applications that need fast, consistent performance at scale without managing database servers. It is designed to deliver single-digit millisecond performance whether you have 100 users or 100 million.

A simple memory rule:

• RDS/Aurora = Relational database (SQL).
• DynamoDB = NoSQL key-value/document database.

---

1) What is Amazon DynamoDB?

Amazon DynamoDB is a serverless, fully managed, distributed NoSQL database.

• Serverless: You don't provision servers, patch software, or worry about disk space.
• Fully Managed: AWS handles all the infrastructure, maintenance, and high availability.
• Distributed: Your data is automatically spread across multiple servers and Availability Zones for durability.

---

2) How it differs from Relational Databases

DynamoDB is fundamentally different from services like RDS or Aurora:

• No Joins: DynamoDB does not support SQL JOIN operations.
• Schema-less: While every item needs a primary key, other attributes can vary from one record to the next.
• Scaling: It scales horizontally to handle virtually unlimited traffic without performance degradation.
• Design: You design your table based on your access patterns (how you query) rather than just how the data is related.

---

3) The Data Model: Tables, Items, and Attributes

DynamoDB uses a simple hierarchy to organize data:

Tables

A collection of data records. (e.g., a Users table).

Items

A single record within a table. This is similar to a "row" in a relational database.

Attributes

A data field within an item. This is similar to a "column."

---

4) The Primary Key

Every item in a table must be uniquely identified by a Primary Key. There are two types:

1. Simple Primary Key (Partition Key): A single attribute (e.g., UserID).
2. Composite Primary Key (Partition Key + Sort Key): Two attributes used together (e.g., UserID + OrderDate). This allows you to store multiple items for the same user and sort them by date.

---

5) Capacity Modes

How do you pay for DynamoDB? You choose a capacity mode:

On-Demand Mode

• Pay-per-request: You only pay for what you use.
• Automatic Scaling: Handles unpredictable traffic spikes instantly.
• Best For: New apps, unknown traffic, or "spiky" workloads.

Provisioned Mode

• Defined Capacity: You specify exactly how many reads/writes per second you want.
• Cost Predictable: Good for steady, predictable traffic.
• Best For: Established apps with consistent usage.

---

Micro-activity 1: SQL or NoSQL?

Micro-activity 2: Capacity Modes

---

Summary

Amazon DynamoDB is the workhorse of modern, serverless AWS applications. It provides incredible speed and scale by trading away the complexity of SQL joins for a simpler, high-performance key-value model.

---

Knowledge Check

Next lesson

Lesson 4.15: Redshift and ElastiCache