AWS Knowledge Hub

English

Tiếng Việt

English

Tiếng Việt

Appearance

Amazon DynamoDB ​

Amazon DynamoDB is a fully managed NoSQL database service that provides fast and predictable performance with seamless scalability.

Overview ​

DynamoDB is a serverless, key-value and document database that delivers single-digit millisecond performance at any scale. It's built for applications that need consistent, single-digit millisecond latency.

Key Features ​

  • Serverless: No servers to manage, automatic scaling
  • Fast Performance: Single-digit millisecond latency
  • Scalable: Handles millions of requests per second
  • Fully Managed: Automatic backups, patching, and monitoring
  • Global Tables: Multi-region, multi-active replication
  • Streams: Real-time data capture and processing
  • On-Demand Mode: Pay-per-request pricing
  • Point-in-Time Recovery: Automatic backups with PITR

Core Concepts ​

Tables ​

Collections of items. Tables are schema-less except for the primary key.

Items ​

Collection of attributes. Each item is uniquely identified by its primary key.

Attributes ​

Name-value pairs. DynamoDB is schema-less, so items can have different attributes.

Primary Key ​

Uniquely identifies each item in a table. Can be:

  • Partition Key: Simple primary key
  • Partition Key + Sort Key: Composite primary key

Secondary Indexes ​

  • Global Secondary Index (GSI): Can have different partition and sort keys
  • Local Secondary Index (LSI): Same partition key, different sort key

Data Models ​

Key-Value ​

Simple key-value pairs, suitable for simple data structures.

Document ​

Structured documents (JSON), supports nested data structures.

Capacity Modes ​

On-Demand ​

  • Pay per request
  • No capacity planning
  • Automatic scaling
  • Suitable for unpredictable workloads
  • Higher cost per request

Provisioned ​

  • Specify read and write capacity units
  • Cost-effective for predictable workloads
  • Auto scaling available
  • Reserved capacity options

Access Patterns ​

Query ​

Retrieve items by partition key (and optionally sort key). Very efficient.

Scan ​

Examine every item in a table. Less efficient, use sparingly.

GetItem ​

Retrieve a single item by its primary key. Most efficient for single items.

Batch Operations ​

  • BatchGetItem: Retrieve up to 100 items
  • BatchWriteItem: Write up to 25 items

Advanced Features ​

DynamoDB Streams ​

  • Time-ordered sequence of item-level changes
  • Enables event-driven architectures
  • Integrates with Lambda, Kinesis
  • Real-time processing capabilities

Global Tables ​

  • Multi-region, multi-active replication
  • Automatic conflict resolution
  • Eventual consistency across regions
  • Disaster recovery and low latency

Transactions ​

  • ACID transactions across multiple items
  • All-or-nothing operations
  • Up to 25 items or 4 MB
  • Conditional writes and deletes

TTL (Time To Live) ​

  • Automatically delete items after expiration
  • No additional cost
  • Useful for session data, temporary records
  • Background deletion process

Security ​

  • Encryption at Rest: AWS KMS integration
  • Encryption in Transit: SSL/TLS connections
  • IAM Integration: Fine-grained access control
  • VPC Endpoints: Private connectivity
  • Point-in-Time Recovery: Automatic backups

Use Cases ​

  • Web Applications: User sessions, shopping carts
  • Gaming: Player data, leaderboards
  • IoT: Device state, telemetry data
  • Mobile Apps: User profiles, app data
  • Real-Time Applications: Live leaderboards, chat
  • Content Management: Metadata, content catalogs

Best Practices ​

  • Design for Access Patterns: Model data based on query patterns
  • Use Composite Keys: Enable flexible querying
  • Avoid Scans: Use queries and GSIs instead
  • Use GSIs Wisely: Balance query flexibility with cost
  • Implement Exponential Backoff: Handle throttling gracefully
  • Use Batch Operations: Reduce round trips
  • Monitor Metrics: Track consumed capacity and throttles
  • Use TTL: Automatically clean up expired data
  • Enable Point-in-Time Recovery: For critical data
  • Use Global Tables: For multi-region applications

Performance Optimization ​

  • Right-Size Capacity: Match provisioned capacity to actual usage
  • Use On-Demand for Variable Workloads: Automatic scaling
  • Optimize Item Size: Smaller items = better performance
  • Use Projection Expressions: Retrieve only needed attributes
  • Implement Caching: Use DAX or ElastiCache
  • Monitor Throttling: Adjust capacity or use on-demand
  • Use Parallel Scans: For large table scans

Cost Optimization ​

  • Use on-demand for unpredictable workloads
  • Use provisioned for predictable, steady workloads
  • Right-size capacity units
  • Use reserved capacity for long-term workloads
  • Enable TTL to reduce storage costs
  • Compress large items
  • Use sparse indexes
  • Monitor and optimize unused capacity

Limits ​

  • Item Size: 400 KB maximum
  • Table Size: Unlimited
  • Throughput: On-demand scales automatically
  • GSIs per Table: 20 (soft limit, can be increased)
  • LSIs per Table: 5
  • Attribute Values: 400 KB per attribute value