Home / Companies / Cube / Blog / Post Details
Content Deep Dive

Introducing Cube Store: High concurrency and sub-second latency for any database

Blog post from Cube

Post Details
Company
Date Published
Author
Igor Lukanin
Word Count
1,033
Language
English
Hacker News Points
-
Summary

Cube.js has released its operational analytics platform, Cube Store, which allows developers to build analytical APIs and semantic layers over any SQL-compliant database or data warehouse with sub-second latency and high concurrency capabilities. Cube Store is a custom pre-aggregation storage layer designed to resolve issues with traditional databases used for pre-aggregations, providing a performant solution with guaranteed high concurrency and low latency performance optimizations for analytical queries. It uses a distributed query engine architecture, Apache Arrow for efficient in-memory data structures, and DataFusion as a query execution framework, written entirely in Rust for optimal performance. Cube Store is now generally available and recommended for use in production, offering benefits such as data federation via cross-database joins, and can be easily scaled horizontally with ease.