May 2024 Summaries

Cube has introduced query pushdown, a new feature that enhances its SQL API's capabilities to support connecting semantic layers to popular BI tools like Power BI, Tableau, Thoughtspot, Sigma, Superset/Preset, and Metabase. Query pushdown enables Cube to understand and execute complex SQL queries generated by these tools. The feature uses Apache DataFusion and egg to analyze incoming SQL queries and find the best query plan for execution. Query pushdown is currently in public preview and will become a generally available feature soon.

RisingWave, an innovative distributed streaming database, is now a supported data source within Cube. It enables real-time data processing and management using Postgres-style SQL syntax. RisingWave can be scaled horizontally and vertically, and supports inflight joins, filtering, aggregations, and other transformations. Users can combine the streaming capabilities of RisingWave with Cube's universal semantic layer to build custom real-time data applications for various sectors such as finance, manufacturing, telecom, and utilities. To use RisingWave as a data source in Cube, simply select the Postgres data source type and enter your credentials.

Cube Cloud offers a two-level caching system, including an in-memory cache and pre-aggregations, which can significantly improve query performance and provide higher concurrency. Load testing results show that Cube Store outperforms cloud data warehousing solutions in terms of peak Requests Per Second (RPS) and response times under various scenarios. As dataset and user numbers grow, Cube Cloud's caching capabilities address the latency issues often faced by application databases and inconsistent response times from cloud data warehousing solutions.

Cube has introduced new advancements in its Cube Cloud, including a native AI API and Chart Prototyping, aimed at accelerating the development of Generative AI (GenAI) and embedded analytics solutions. The universal semantic layer provided by Cube's cloud enhances context and governance for LLMs, reducing errors and hallucinations. The new AI API simplifies application-LLM integrations and enables users to create AI agents, chatbots, copilots, and Slack apps using natural language interactions. Chart Prototyping allows developers to create and test charts within Cube Cloud's Playground 2.0, speeding up the development process for embedded analytics. These features are designed to help organizations unlock the full potential of their data and build secure, accurate, and cost-effective data experiences.

Cube's new AI API is now available for users of Cube Cloud Premium and higher tiers. The API includes Retrieval Augmented Generation (RAG) and prompt engineering, allowing developers to build natural language capabilities with a turnkey solution starting with OpenAI. The AI API leverages the company's deep knowledge of its universal semantic layer, APIs, and metadata, enabling users to focus on developing the best AI agents and chatbot experiences for their customers without needing to build their own RAG, prompt engineering, or vector store. The semantic layer provides context and constrains what the LLM can answer with, making it a reliable middleware between LLMs and data.

Cube introduces Chart Prototyping, a new feature that accelerates the development of front-end applications by generating code for querying data via its REST API and visualizing it on charts. This tool supports various chart types and can be customized to connect with different data visualization tools or build bespoke data applications. Currently, Chart Prototyping generates TypeScript code for React applications using Chart.js as a charting library. Users can try this feature in Cube Cloud's Playground by creating a query, visualizing it on a chart, and downloading the generated code. Feedback is welcome through their Slack community.

The democratization of data has led to an increase in the use of traditional business intelligence (BI) platforms, but this growth does not necessarily translate into more business value. Modern analytics platforms often overwhelm users with a large volume of data sources and dashboards, leading to frustration and decreased productivity. Curated data experiences through embedded analytics can help address these issues by providing relevant insights within the context of an employee's workflow. Embedded analytics enhances internal decision-making, boosts employee productivity and engagement, and optimizes business processes by integrating data-driven insights directly into operational tools. By leveraging platforms like Cube Cloud, organizations can build customized embedded analytics solutions to drive growth, innovation, and continuous improvement across all aspects of the business.