DynamoDB at a Glance

A serverless NoSQL database

DynamoDB is a fully managed NoSQL database with single-digit-millisecond reads and writes. You design a table around a partition key (and optional sort key) and query by those keys. Capacity can be on-demand (pay per request) or provisioned.

Put and query items

Design for your access patterns first: a query by partition key is fast, while a full table scan is slow and costly.

aws dynamodb put-item --table-name Orders \
  --item '{"customerId":{"S":"c42"},"orderId":{"S":"o1001"},"total":{"N":"499"}}'

aws dynamodb query --table-name Orders \
  --key-condition-expression "customerId = :c" \
  --expression-attribute-values '{":c":{"S":"c42"}}'

Choose a partition key with many distinct values so load spreads evenly. Use DynamoDB when access patterns are known; use RDS when you need flexible joins and ad hoc queries.

Quick check

Quick check: What does a DynamoDB query use to find items fast?

• The partition key (and sort key)
• A full table scan
• SQL joins
• The AMI id