This toolkit simplifies the deployment and sharing of Databricks Lakehouse demos, targeting users who want to quickly implement and showcase end-to-end Lakehouse solutions. It provides a curated list of demos covering data engineering, ML, and BI use cases, enabling rapid setup and exploration of Databricks capabilities.

How It Works

The dbdemos library acts as an installer and orchestrator for pre-packaged demo bundles. It leverages a configuration file (bundle_config) within each demo to define notebooks, Delta Live Table (DLT) pipelines, DBSQL dashboards, and ML models. Upon installation, dbdemos can automatically provision necessary Databricks resources like clusters and DLT pipelines, and configure dashboards, streamlining the setup process.

Quick Start & Requirements

• Install via pip: %pip install dbdemos
• Requires Databricks cluster creation, DLT pipeline creation, and DBSQL dashboard/query creation permissions.
• Unity Catalog (UC) must be available for UC-specific demos.
• See demo video and explore demos at dbdemos.ai.

Highlighted Details

• Automates the setup of DLT pipelines, DBSQL dashboards, and ML models.
• Dynamically updates demo links to point to created resources for easy navigation.
• Supports templating ({{ }}) for user-specific configurations during installation.
• Provides a framework for developers to contribute new demo bundles.

Maintenance & Community

• Issues can be reported via GitHub issues.
• Databricks does not offer official support; issues are addressed on a best-effort basis.

Licensing & Compatibility

• Licensed under the LICENSE file.
• Usage implies consent for data collection for product improvement; can be disabled by setting Tracker.enable_tracker to False.

Limitations & Caveats

Users are responsible for managing resources created by dbdemos and their associated consumption, even with auto-termination features. Databricks does not provide official support for dbdemos or its assets.