Читать книгу Official Google Cloud Certified Professional Data Engineer Study Guide - Dan Sullivan - Страница 38

For the purposes of choosing a storage technology, it is helpful to consider how data is structured. There are three widely recognized categories:

 Structured

 Semi-structured

 Unstructured

These categories are particularly helpful when choosing a database.

Structured Data Structured data has a fixed set of attributes that can be modeled in a table of rows and columns.

Semi-Structured Data Semi-structured data has attributes like structured data, but the set of attributes can vary from one instance to another. For example, a product description of an appliance might include length, width, height, weight, and power consumption. A chair in the same catalog might have length, width, height, color, and style as attributes. Semi-structured data may be organized using arrays or sets of key-value pairs.

Unstructured Data Unstructured data does not fit into a tabular structure. Images and audio files are good examples of unstructured data. In between these two extremes lies semi-structured data, which has characteristics of both structured and unstructured.