Summary of Atscale - Data Maturity - Module 4 Combining data integration styles

00:00:00 - 00:05:00

This video discusses the different types of data integration styles and provides examples of each. Data movement is the most efficient style, but it requires copies of data that are eventually stale. ETL is a modern approach that does the same data movement in a single step. Data virtualization offers the advantage of near real-time data, but it requires virtualization technology. Atscale recommends using a mix of data integration styles in order to best suit the specific needs of your application.

• 00:00:00 The video discusses the difference between data extracts and data integration styles, and provides examples of each. The three main integration styles are ETL, data virtualization, and data movement. Data movement is the most efficient style, but it requires copies of data that are eventually stale. ETL is a modern approach that does the same data movement in a single step. Data virtualization offers the advantage of near real-time data, but it requires virtualization technology.
• 00:05:00 Data integration is a process of combining data from different sources in order to make it more usable and efficient. There are three stages of data integration: bronze, silver, and gold. Bronze is the stage where data is ingested and landed in its raw form. Silver is where you might enhance data with additional metadata and master data. Gold is the stage where you may add features from your machine learning processes and aggregations to support faster queries. Atscale recommends using a mix of data integration styles in order to best suit the specific needs of your application. Data virtualization can be used to quickly access and use data before it is moved through a data pipeline, and semantic layer access allows users to consume data without understanding its origin or meaning.