database

overview about DB in AWS

SQL vs NoSQL

	SQL	NoSQL
optimal workloads	transactional, OLTP (Online Transaction Processing), OLAP (Online Analytical processing)	no structure, semi-structured data
data model	tables (row, col), indexes, relationship between tables.	key-value, document, in-memory, graph
ACID properties	Atomicity Consistency Isolation Durability	DynamoDB support ACID, high throughput, low latency, horizontal scaling
performance	depend on: disk, queries, indexes, table structure.	hardware cluster size, network latency, calling application.
scale	scale up: increasing compute capabilities scale out: adding replicas.	scale out: distributed architecture.
APIs	SQL queries	Object-based APIs

Services

• Relational DB

  • Traditional

    • Amazone RDS

    • Amazon Aurora

  • Warehouse

    • Amazon Redshift

• Non-relational DB

  • Key-value: DynamoDB

  • Document: DocumentDB

  • In-memory: ElastiCache

  • Graph: Neptune

Use cases

Best practices

• Keep only a subset of the data in RDS

• Move CLOBs and BLOBs to S3

• Move historical data to S3

• Move high velocity frequently written data to DynamoDB

• Do incremental backups

• Communicate within the same AZ

• Purchase reserved instances when possible.