Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

🏠 Back to Blog

What is Database ?

  • A database is data-store that store semi-structured and structured data
  • A database is more complex stores because it requires using formal design and modeling techniques
  • Database types:
    • Relational Database
      • Structured data represents tabular data (tables,rows and columns)
    • Non-Relational Database
      • Semi-Structured that may or may not represent tabular data
  • Set of functionality:
    • query
    • modeling strategies to optimize retrieval for different use cases
    • control over the transformation of the data into useful data structures or reports Database

What is Data Warehouse ?

  • Relational Database : designed for analytic workloads and a column-oriented data-store
  • Companies will have terabytes and millions of rows of data
  • Data warehouses generally perform aggregation
    • aggregation is is grouping data eg. finding a total or average
    • Data warehouses are optimised around columns since they need quickly aggregate column data
  • Data warehouses are generally designed be HOT
    • HOT means they can return queries very fast even though they have vast amounts of data
  • Data warehouses are infrequently accessed
    • intended for real time reporting but maybe once or twice a day or once a week to generate business or user reports
  • Data Warehouse needs to consume data from a relational database on a regular basis

What is a Key Value Store ?

  • A key-value database is a type of non-relational database (NoSQL) that uses a simple key-value method to store data
    • Stores a unique key alongside a value
    • will interpret this data resembling a dictionary
    • can resemble tabular data, it does not have to have the consistent columns per row -Due to simple design so they can scale well beyond a relational database

What is a document database ?

  • Document store - a NOSQL database that stores documents as its primary data-structure

    • it could be an XML but more commonly is JSON or JSON-like

    • they are sub-class of key/value stores

NOSQL Database Services

  • DynamoDB

    • a serverless NOSQL key/Value and document database
    • designed to scale to billions of records with consistent data return in at least a second- millisecond latency
    • It is AWS’s flagship database service meaning it is cost-effective and very fast
    • DAX cluster for read cache, microsecond read latency
    • Event Processing: DynamoDB Streams to integrate with AWS Lambda, or Kinesis Data Streams
    • Global Table feature: active-active setup
    • Automated backups up to 35 days with PITR (restore to new table), or on-demand backups
    • Export to S3 without using RCU within the PITR window, import from S3 without using WCU
    • Great to rapidly evolve schema
    • It is a massively scalable database
    • Usecases: Serverless applications development (small documents 100s Kb), distributed serverless cache

  • DocumentDB

    • A NOSQL document database that is “MongoDB compatible”
    • MongoDB is very popular NOSQL among developers there were open-source licensing issues around using open-source MongoDB , so aws got aorund it by just building their own MongoDB database
    • when you want a MongoDB database

  • Amazon KeySpaces

    • A fully managed Apache Cassandra database
    • Cassandra is an open-source NOSQL key/value database similar to DynamoDB in that is columnar store database but has some additional functionality
    • when you want to use Apache Casandra

Relational Database Service

  • Relational Database Services (RDS)

    • supports multiple SQL engines
    • Relational is synonymous with SQL and Online Transactional Processing (OLTP)
    • most commonly used type of database among tech companies and start ups
    • RDS supports the following SQL Engines:
      • MYSQL - Most popular open source SQL database that was purchased and now owned by Oracle
      • MariaDB - When Oracle bought MYSQL. MariaDB made a fork (copy) of MYSQL was made under a different open-source license
      • Postgres (PSQL) - Most popular open-source SQL database among developers. Has rich-features over MYSQL but at added complexity
      • Oracle - Oracle’s proprietary SQL database. Well used by Enterprise companies. Have to buy a license to use it
      • Microsoft SQL Server - Microsoft’s proprietary SQL database. Have to buy license to use it
      • Aurora - Fully managed database
        • Aurora
          • fully managed database,
          • database of either MYSQL (5X faster) and PSQL (3X faster) database
          • When you want a highly available, durable, scalable and secure relational database for Postgres or MySQL then Aurora is correct fit
      • Aurora Serverless - serverless on-demand version of Aurora. - When you want “most” of the benefits of Aurora but can trade to have cold-starts or you don’t have lots of traffic demand
      • RDS on VMware - allows you to deploy RDS supported engines to on-premise data center. - datacenter must be using VMware for server virtualization - when you want databases managed by RDS on your own datacenter

Other Database Services

  • RedShift
    • petabyte-size data-warehouse
    • Data warehouses
      • are for Online Analytical Procesing (OLAP)
      • can be expensive because they are keeping data “hot”
      • “HOT” means we can run a very complex query and a large amount of data and get that data very fast
      • Usage: when you want to quickly generate analytics or reports from a large amount of data
  • ElasticCache
    • a managed database of the in-memory and caching open-source databases
    • Redis or Memcached
    • Usage: when you want to improve the performance of application by adding a caching layer in-front of web-server or database
  • Neptune
    • a managed graph database
    • Data is represented in interconnected nodes
    • Usage: when you need to understand the connections between data eg. Mapping Fraud Rings or Social Media Relationships
  • Amazon Timestreams
    • a fully managed time series database
    • Related to Devices that send lot of data that are time-sensitive such as IOT devices
    • Usage: When you need to measure how things change over time
  • Amazon Quantum Ledger Database
    • a fully managed ledger database that provides transparent, immutable and cryptographically variable transaction logs
    • Usage: when you need to record history of financial activities that can be trusted
  • Database Migration Service
    • a database migration service
    • Can migrate from:
      • On-premise database to AWS
      • from two database in different or same AWS accounts using SQL engines
      • from a SQL to NOSQL database