Note_Tech

All technological notes.

---

Project maintained by simonangel-fong Hosted on GitHub Pages — Theme by mattgraham

AWS - NoSQL

Back

• AWS - NoSQL
  • NoSQL Databases
    • A Key/Value Store
    • Document Store
  • NoSQL Database Services
    • DynamoDB - Serverless, Low latency retrieval
      • Global Tables - low latency, Active-Active, multiple-regions
    • DynamoDB Accelerator(DAX) - In-memory Cache
    • DocumentDB - MongoDB
    • Amoazon Keyspaces - Cassandra

---

NoSQL Databases

• NoSQL = non-SQL = non relational databases
• NoSQL databases are purpose built for specific data models and have flexible schemas for building modern applications.
• Benefits:
  • Flexibility: easy to evolve data model
  • Scalability: designed to scale-out by using distributed clusters
  • High-performance: optimized for a specific data model
  • Highly functional: types optimized for the data model

• Examples: Key-value, document, graph, in-memory, search databases

• NoSQL data example: JSON
  • JSON = JavaScript Object Notation
  • JSON is a common form of data that fits into a NoSQL model
  • Data can be nested
  • Fields can change over time
  • Support for new types: arrays, etc…

---

A Key/Value Store

• A key-value database is a type of non-relational databse (NoSQL) that uses a simple key-value method to store data.

  • Key values stores are dumb and fast.
  • generally lack feactures like:
    • relationships
    • indexes
    • Aggregation

---

Document Store

• Document Store
  • a NOSQL database that stores documents as its primary data structure.
  • A document could be an XML but commonly is JSON or JSON-Like
  • Document stores are sub-class of Key/Value stores.

---

NoSQL Database Services

DynamoDB - Serverless, Low latency retrieval

• DynamoDB

  • NoSQL database - not a relational database
  • DynamoDB is a key/value database
  • Single-digit millisecond latency – low latency retrieval
  • Scales to massive workloads, distributed “serverless” database
• Fully Managed Highly available with replication across 3 AZ
• Millions of requests per seconds, trillions of row, 100s of TB of storage
• Fast and consistent in performance
• Integrated with IAM for security, authorization and administration
• Low cost and auto scaling capabilities
• Standard & Infrequent Access (IA) Table Class
• Can Create table without a database

Global Tables - low latency, Active-Active, multiple-regions

• Make a DynamoDB table accessible with low latency in multiple-regions
• Active-Active replication (read/write to any AWS Region)

---

DynamoDB Accelerator(DAX) - In-memory Cache

• Fully Managed in-memory cache for DynamoDB
• 10x performance improvement – single- digit millisecond latency to microseconds latency – when accessing your DynamoDB tables
• Secure, highly scalable & highly available
• Difference with ElastiCache at the CCP level:
  • DAX is only used for and is integrated with DynamoDB,
  • while ElastiCache can be used for other databases

---

DocumentDB - MongoDB

• Aurora is an “AWS-implementation” of PostgreSQL / MySQL …
• DocumentDB is the same for MongoDB (which is a NoSQL database)
• MongoDB is used to store, query, and index JSON data
• Similar “deployment concepts” as Aurora
• Fully Managed, highly available with replication across 3 AZ
• DocumentDB storage automatically grows in increments of 10GB, up to 64 TB.
• Automatically scales to workloads with millions of requests per seconds

---

Amoazon Keyspaces - Cassandra

• Cassandra

  • an open-souce NoSQL key/value database.

• Amoazon Keyspaces

  • a fully managed Apache Cassardra database.
  • For users who want to use Apache Cassandra

---

TOP