Back to main

Video: SQL vs NoSQL Explained

https://www.youtube.com/watch?v=ruz-vK8IesE

What is SQL?

• Structured Query Language

• Allows CRUD operations through universal language

• 4 key components

• Structure

• Tables w/ rows (individual entities/elements) and columns (types)

• Data can have constraints

• Consistency across contents of table

• Relations

• Storage

• Storage pattern is concentrated

• 1 node that contains entire copy of data, not partitioned

• Scale

• Vertical (better machine) - more memory, faster CPU, better hard disk, etc

• Upper ceiling: physical limitation

• Horizontal

• Adding more machines

• Distributing data across multiple nodes

• Access

• Raw SQL

• Require direct database connection

• Object relational mappers

What is NoSQL?

• Anything that is non-relational

• Many implementations: table, doc, graph, etc

• Built to scale with high performance, but queries are less flexible

• 4 key components

• Structure

• Tables

• Documents (JSON)

• Graphs

• Rely on Key Value Store - need to know key when performing query

• Storage

• Hashing Input f(x)

• Depending hashing value, it is distributed onto different partitions

• Scale

• Add more partitions

• Access

• REST APIs

• Handle specific endpoint that has specific functionality

• CRUD in Vendor specific language

When to use what?

• SQL

• When your access patterns aren't defined

• Not sure yet how your business wants to use data

• When you want to perform flexible queries

• When you want to perform relational queries

• If you want query to make multiple hops between tables

• When you want to enforce field constraints

• When you want to use a well documented access language (SQL)

• NoSQL

• When your access pattern is defined

• When your primary key is known

• When your data model fits (graphs)

• When you need high performance and low latency