Skip to main content
Datazone Views

Overview

Views enable you to transform your datasets into optimized relational database structures with advanced configurations like partitioning, primary keys, and ordering. This feature bridges the gap between Datazone’s lakehouse architecture and traditional relational database capabilities, providing enhanced performance and flexibility for your data access patterns. With Views, you can replicate datasets or create custom aggregations using SQL queries across multiple datasets and even other views. The resulting views are accessible through Datazone’s SQL interface, endpoints, and intelligent apps.

View Types

  • Materialized View: Stores the query results in the database, so queries are very fast and consistent until refreshed. Best for dashboards and reports you check often.
  • Non-Materialized View: Runs the query every time you access it, always showing the latest data. Good for real-time needs or when you want to save storage.
Materialized vs Non-Materialized Views
If you create non-materialized views, keep in mind that query performance may vary based on the complexity of the underlying SQL and the size of the datasets involved. For frequently accessed views, consider using materialized views to enhance performance.

Let’s Create a View

With the intuitive interface, you can easily create and manage views to suit your data needs.
  1. Go to the Views page by clicking on the Views tab in the sidebar.
  2. Click on the Create button.
  3. Choose between creating a Materialized or Non-Materialized view.
  4. Select whether it’s a replicate of a dataset or a custom SQL query.
  5. Fill the optimization customization options as needed.
  6. Check the preview and create the view. 🚀

Why I Should Use Views

Let’s say you have a large dataset that is frequently queried for reporting purposes. By creating a materialized view with appropriate partitioning and primary keys, you can significantly speed up query performance, reduce load times, and improve the overall user experience for your data consumers. In the following scenario, you have a sales dataset that is regularly queried to generate monthly sales reports.
Materialized vs Non-Materialized Views
By creating a materialized view that aggregates sales data by category or day of week, you can optimize the performance of these queries, making report generation faster and more efficient.

Next Steps

I