Denormalization

• improve read performance at the expense of write performance.
• write redundant copies of data to multiple tables.
  • this allows most complex join operations to be avoided.
    • this is important because when we split up our databases with database-sharding or database-federation, joins become more difficult to do.
• disadvantages.
  • duplicated data.
    • redundant data needs to be kept in sync.
  • slow writes.

table without id file.inlinks as Backlinks
where file.name = this.file.name

References.

• https://en.wikipedia.org/wiki/Denormalization

Categories:: database