Day 27 - Amazon RDS & Aurora

Date: 2025-10-14 (Tuesday)
Status: “Done”

Lecture Notes

RDBMS vs NoSQL

RDBMS

  • An RDBMS stores data in related tables (rows/columns), enforces integrity constraints, uses SQL, and provides ACID guarantees.
  • Popular engines: Oracle, MySQL, SQL Server, PostgreSQL, IBM Db2.

image

NoSQL Overview

  • NoSQL systems target un/semistructured data with high scalability and performance.
  • Types:
    • Document (MongoDB, CouchDB)
    • Key–Value (Redis, DynamoDB)
    • Column-Family (Cassandra, HBase)
    • Graph (Neo4j, Amazon Neptune)
  • Traits: schema flexibility, horizontal scaling, big-data friendliness, CAP-oriented designs.

image

RDBMS vs. NoSQL (high-level)

image

OLTP vs. OLAP

  • OLTP: many small, concurrent transactions; normalized data; short queries; index-heavy.
  • OLAP: complex analytics over large historical datasets; star/snowflake schemas; read-heavy.

image

Amazon RDS & Aurora

Amazon Relational Database Service (RDS)

Managed relational databases that simplify provisioning, patching, backups, and HA.

  • Supported engines: MySQL, PostgreSQL, MariaDB, Oracle, SQL Server, Amazon Aurora.
  • Key features: automated backups/patching, easy scaling, Multi-AZ high availability, encryption & VPC/IAM/SSL security.
  • Deployment options:
    • Single-AZ
    • Multi-AZ (synchronous standby in another AZ)
    • Read Replicas for scaling reads

image image

RDS Features:

  • Automated Backups: Point-in-time recovery up to 35 days
  • Manual Snapshots: User-initiated backups
  • Multi-AZ: Automatic failover for high availability
  • Read Replicas: Scale read workloads (up to 15 replicas)
  • Parameter Groups: Database configuration management
  • Option Groups: Additional features (e.g., Oracle Advanced Security)

Amazon Aurora

Cloud-native, MySQL/PostgreSQL-compatible relational database re-architected for AWS.

image

  • Highlights:
    • Up to ~5× MySQL / ~3× PostgreSQL performance (typical benchmarks)
    • Storage auto-scales to 128 TB
    • Six-way replication across three AZs; self-healing storage
    • Aurora Serverless (on-demand capacity)
    • Global Database for low-latency multi-region

image image

Aurora Features:

  • Aurora Replicas: Up to 15 read replicas with sub-10ms lag
  • Aurora Serverless: Auto-scaling compute capacity
  • Aurora Global Database: Cross-region replication < 1 second
  • Aurora Backtrack: Rewind database to specific point in time
  • Aurora Parallel Query: Faster analytics on current data
  • Aurora Machine Learning: Native ML integration

Aurora vs RDS:

Feature Aurora RDS
Performance 5x MySQL, 3x PostgreSQL Standard
Storage Auto-scaling to 128 TB Manual scaling
Replicas Up to 15 Up to 5 (MySQL)
Failover < 30 seconds 1-2 minutes
Backtrack Yes No

Hands-On Labs

Lab 05 – Amazon RDS & EC2 Integration (Part 2)

  1. Create EC2 Instance → 05-3
  2. Create RDS Database Instance → 05-4
  3. Application Deployment → 05-5
  4. Backup and Restore → 05-6
  5. Clean Up Resources → 05-7

Lab 43 – AWS Database Migration Service (DMS) (Part 1)

  1. EC2 Connect RDP Client → 43-01
  2. EC2 Connect Fleet Manager → 43-02
  3. SQL Server Source Config → 43-03
  4. Oracle Connect Source DB → 43-04
  5. Oracle Config Source DB → 43-05
  6. Drop Constraint → 43-06