August 2022 Summaries

We started Cube as an open-source analytics framework in 2019, written in JavaScript, but since then we have made significant changes, including moving from npm to Docker for distribution and increasing our use of Rust in the codebase. Our SQL API was recently released entirely in Rust and we plan to migrate the rest of the codebase to it over time. We also plan to expand support to YAML and Python later this year. The name change reflects Cube's evolution from a JavaScript-based analytics framework to a headless BI platform, but no action is required from our customers as the change is purely cosmetic.

The article explains how to use Firebolt and Cube to generate insights from Wikipedia page view logs, with the goal of identifying outliers in the data. The process involves creating a Firebolt account, setting up a database and table structure, ingesting data from S3 into Firebolt, calculating time-window aggregations, searching for outliers, and storing the results in a separate fact table. Additionally, Cube is used to encapsulate queries, provide access control, and deliver the data via its REST API to end users of a web application built with React. The article also discusses the technology behind Firebolt's high performance and how it can be connected to Cube for a seamless data analytics experience.

Embedded analytics refers to the integration of data analytics and visualization features directly within a user's existing workflow, allowing for real-time insights without the need to switch between different applications. This approach enhances user experience and increases the adoption of tools by providing seamless, user-centric designs that are particularly beneficial for both internal and external users. Embedded analytics can lead to increased revenue and competitive advantage for businesses through improved user experience, tiered service offerings, and upselling opportunities. Internally, it boosts productivity by eliminating cumbersome data processes and empowering data-driven decision-making across all levels of expertise. The key distinction between embedded analytics and traditional business intelligence lies in the former's ability to offer real-time data insights within the operational application itself, streamlining the analytics process and saving significant time. Cube, a headless business intelligence tool, facilitates the development of embedded analytics by providing a centralized semantic layer for accurate metrics and secure data access, making it an efficient solution for building customizable and performant analytics applications.