May 2022 Summaries

Cube has helped thousands of companies consume data from any source, organize it into consistent definitions, and use it with data applications since its open-source launch in 2019. The company aims to make Cube's power more accessible to a broader range of users, including analysts and business users who don't typically work with code. To achieve this, Cube has updated its SQL API to enable integration with popular business intelligence tools like Tableau, Superset, and Power BI. This update allows users to connect any data visualization and BI application to Cube via the Postgres SQL interface, providing consistent data definitions across an organization. With this change, Cube reduces the difficulty of exploring data with collaborative data tools and notebooks, centrally manages access to up-stack data stores, and minimizes bandwidth costs and degraded performance by caching queries and aggregate tables. The new SQL API is now available in both the open-source Cube tool and Cube Cloud, a managed service hosted within select regions of Google Cloud Platform and Amazon Web Services.

Ternary, a FinOps platform for Google Cloud customers, addresses the challenge of managing large volumes of cost-related data by implementing a 3-query approach to enable multidimensional analysis and optimize cloud spending. This method involves three key steps: aggregating overall cost totals over a specific period, identifying the top N combinations of projects and categories with the highest spending, and using this data to generate detailed breakdowns while accounting for the remaining data in an "other" column. This approach ensures that users receive a comprehensive view of their expenses without exceeding data limits, improving data visualization and decision-making capabilities. While this method increases the number of queries and potential costs, especially when using on-demand pricing for data queries, it provides a more complete depiction of user data, balancing performance with the cost of implementation.

Materialize is a database that maintains the results of a query as a materialized view, making it great for scenarios where complex join and transformation capabilities are needed but data needs to be served quickly. Cube is a headless business intelligence platform that provides data modeling, access control, caching, and APIs on top of any data warehouse or query engine, including Materialize. Integrating Cube with Materialize allows for the delivery of consistent data to BI tools, data apps, and notebooks. Together, they enable real-time data analysis and provide a "live read API" for entire businesses, addressing performance problems associated with traditional data warehouses and connecting systems to data warehouses in a more efficient way.

The Cube Core v0.30 release brings a significant number of new features and improvements across various platforms, including support for dbt metrics with Cube, Databricks export bucket feature, Materialize databases, QuestDB databases, and many others. The Developer Playground has also seen quality-of-life improvements, while the Cube SQL API has been vastly improved since its launch in v0.29. Breaking changes include deprecating CUBEJS_EXTERNAL_DEFAULT and CUBEJS_SCHEDULED_REFRESH_DEFAULT environment variables. Notable fixes and additions include support for ARM64 Linux, MySQL compatibility with information_schema, and various performance enhancements. The release also includes new features for AWS Athena, BigQuery, Databricks, JDBC, Materialize, MS SQL, Postgres, QuestDB, Snowflake, and Server platforms. Additionally, improvements have been made to the Playground, Client libraries, and documentation.