Skip to content

8-Databases.md

Latest commit

 

History

History
216 lines (144 loc) · 6.28 KB

8-Databases.md

File metadata and controls

216 lines (144 loc) · 6.28 KB

Chapter 8: Databases

Chapter Overview

Topics

  • Amazon Relational Database Service (Amazon RDS)
  • Amazon DynamoDB
  • Amazon Redshift
  • Amazon Aurora

Lab

  • Build Your DB Server and Interact with Your DB Using an App

Chapter Objectives

After completing this chapter, you should be able to:

  • Explain Amazon RDS and its core functionality
  • Describe Amazon DynamoDB and its key features
  • Understand Amazon Redshift and Amazon Aurora
  • Launch, configure, and interact with an RDS database
  • Differentiate between managed and unmanaged database services
  • Choose the right AWS database solution for different scenarios

Section 1: Amazon Relational Database Service (RDS)

What is Amazon RDS?

  • Managed relational database service (RDS) for SQL databases
  • AWS handles scaling, fault tolerance, backups, patching, and high availability
  • You focus on application optimization

Supported Engines

  • Amazon Aurora
  • PostgreSQL
  • MySQL
  • MariaDB
  • Oracle
  • Microsoft SQL Server

Key Features

  • Database instance classes (CPU, memory, network)
  • Storage options: Magnetic, General Purpose SSD, Provisioned IOPS
  • High availability via Multi-AZ deployment (primary and standby in different AZs)
  • Read replicas for scaling read-heavy workloads (asynchronous replication)
  • Secure and scalable in a VPC, with subnets and security groups

Pricing & Deployment

  • Clock-hour billing: Pay for running resources
  • Purchase types: On-demand (pay by hour), Reserved Instances (discounts for commitment)
  • Storage: Provisioned (included), backup storage (GB/month)
  • Deployment: Single AZ or Multi-AZ, data transfer in is free, outbound is tiered

Use Cases

  • Web/mobile apps: high throughput, scalable, high availability
  • E-commerce: data security, low cost, managed solution
  • Games: rapid scaling, automatic throughput, monitoring

When to Use RDS

  • Need for complex transactions/queries, medium-high query rates, high durability
  • Not ideal for massive throughput, sharding, or simple GET/PUT NoSQL

Key Takeaways

  • RDS is a managed, scalable, highly available relational database
  • Accessible via console, CLI, and API
  • Supports several major database engines

Section 2: Amazon DynamoDB

Relational vs Non-Relational Databases

Aspect Relational (SQL) Non-Relational (NoSQL)
Data Storage Rows/columns Key-value, document, graph
Schemas Fixed Dynamic
Querying SQL Document/collection focus
Scalability Vertical Horizontal

What is DynamoDB?

  • Fast, flexible, fully-managed NoSQL database
  • Tables, items, and attributes (schema-less)
  • Supports partition keys and sort keys
  • SSD-backed, virtually unlimited storage
  • Scalable read/write throughput, single-digit millisecond latency
  • Automatic replication across chosen AWS Regions

Use Cases

  • Mobile/web/gaming, adtech, IoT
  • Accessible via console, CLI, API

Key Takeaways

  • DynamoDB supports key-value and document models
  • Scalable, low-latency, no limits on table size or throughput
  • Fully managed, runs on SSDs

Section 3: Amazon Redshift

Overview

  • Fully managed, fast, scalable data warehouse service for analytics and big data
  • Columnar storage and parallel processing architecture
  • Leader node and compute nodes (virtual cores, RAM, local disk)
  • Integrates with SQL clients, BI tools, S3, DynamoDB

Features

  • Automatic scaling and monitoring
  • Built-in encryption
  • No downtime for scaling

Use Cases

  • Enterprise data warehouse, analytics, SaaS, big data
  • Experimentation, migration from on-premises, manage as business grows

Key Takeaways

  • Redshift is fast, scalable, and managed for big data analytics
  • Supports columnar storage, parallel processing, automatic monitoring

Section 4: Amazon Aurora

Overview

  • Enterprise-class relational database compatible with MySQL and PostgreSQL
  • Automates provisioning, patching, backup, recovery, failure detection/repair
  • Multiple levels of security

Features

  • High performance, scalability, availability, durability
  • Fully managed, easy migration

Key Takeaways

  • Aurora offers high performance and availability, with compatibility for MySQL/PostgreSQL
  • Fully managed, automated tasks, multiple security layers

Section 5: Choosing the Right Database Tool

Requirement AWS Solution
Enterprise-class relational DB Amazon RDS
Fast, flexible NoSQL Amazon DynamoDB
OS/app features not supported by AWS Databases on EC2
Special cases (analytics, ML, graph) Purpose-built AWS DB services

Section 6: Database Case Studies

Case 1: Data Protection/Management

  • Needs: relational DB for config, unstructured store for metadata, S3 for storage, Glacier for archive
  • Solutions: RDS/Aurora for config, DynamoDB for metadata

Case 2: Shipping Company Migration

  • Needs: move to serverless, legacy Oracle, structured/semistructured data
  • Solutions: Aurora (PostgreSQL/MySQL), DynamoDB for semi-structured

Case 3: Online Payment Processing

  • Needs: high throughput for flash sales, millions of transactions/day, read replicas for scaling
  • Solutions: RDS with read replicas, IAM and KMS for security

Sample Exam Question

Which of the following is a fully-managed NoSQL database service?

Choice Response
A Amazon Relational Database Service (Amazon RDS)
B Amazon DynamoDB
C Amazon Aurora
D Amazon Redshift

Answer:
The correct answer is B.
Keywords: "NoSQL database service".

Resources