June 2021 Summaries

Apache Superset, a robust business intelligence tool, is gaining traction among ClickHouse users for its creative data visualization capabilities. This blog post, originally published by Altinity, details collaborative efforts to enhance Superset's connectivity with ClickHouse, recommending the use of the clickhouse-sqlalchemy driver for optimal performance and security. It outlines three methods for installing Superset: using a Python virtual environment, Docker Compose, and the managed Preset Cloud service. The post provides step-by-step instructions for setting up Superset and connecting it to a ClickHouse database, alongside creating datasets and visualizations without needing SQL expertise. Additionally, it discusses the underlying connectivity architecture, emphasizing the preference for native TCP protocol and the importance of using updated software versions. The blog also highlights the collaborative work between Altinity and Preset to improve ClickHouse support and invites users to explore Superset's extensive capabilities for visualizing data, encouraging community engagement and sharing of innovative use cases.

Apache Superset has integrated several new chart types built on Apache ECharts into its platform, now available in Preset Cloud and on the bleeding edge master branch of Superset's source code. These chart types include Funnel, Force Directed Graph, Circular Graph, Mixed Timeseries, Gauge, Radar, Tree, Pivot Table v2, and Tree Map v2, each designed for specific data visualization needs. Funnel charts are useful for visualizing stages in a pipeline, while force directed and circular graphs effectively showcase many-to-many relationships. Mixed timeseries charts allow the combination of different timeseries objects, and gauge charts illustrate progress towards specific goals in an easily comprehensible manner. Radar charts help visualize metrics across multiple categories, and tree charts excel at depicting hierarchical data structures. The redesigned Pivot Table v2 and Tree Map v2 enhance data visualization capabilities with options like heatmap and bar chart integrations and better handling of categorical data distributions. Most of these charts are expected to feature in the upcoming stable release of Superset 1.2, pending community approval.

The text discusses the integration and benefits of using Trino, a SQL query engine, with Apache Superset for data visualization, particularly highlighting its ability to handle ad-hoc queries across heterogeneous data sources like MongoDB. Originating from a fork of the Presto project and now known as Trino, it serves as a single access point to query multiple data sources without storing data itself, leveraging a connector architecture that allows communication with various databases. The document provides insights into setting up Trino using Docker, outlining the necessary configurations and emphasizing Trino's capability to translate different databases' query languages into standardized ANSI SQL, thereby simplifying data access for analytics teams. It notes the advantage of using Trino for Superset users, enabling them to visualize data from previously unsupported databases like MongoDB, and touches on the ease of configuration and deployment using Docker containers. The text concludes by hinting at further exploration of visualizing Trino-queried data using Superset in an upcoming post.