Getting Started with Cube: Set Up Your First Cube Project in Minutes

  

If you're exploring semantic modeling, analytics APIs, or modern BI infrastructure, Cube is one of the most powerful open-source platforms available today.

 

In this blog post, we’ll walk through how to quickly set up a local Cube project using PostgreSQL as the data source.

 

1. What is Cube?

Cube is a semantic layer and analytics platform that helps developers build consistent, scalable, and high-performance data applications.

 

It sits between your data warehouse/database and frontend applications, enabling:

 

·      Semantic modeling

·      Metrics definitions

·      Query generation

·      Caching and pre-aggregations

·      API access for analytics

·      Integration with BI tools and custom dashboards

 

Cube is widely used for building internal analytics platforms, customer-facing dashboards, and modern BI experiences.

 

2. Prerequisites

Before starting, make sure you have:

 

·      Node.js installed

·      npm installed

·      PostgreSQL database access

 

You can verify installation using:

node -v
npm -v

$node -v
v24.7.0
$
$
$npm -v
11.5.1

   

3. Install Cube CLI

Cube provides a CLI tool that helps bootstrap projects quickly.

 

You can install it globally using:

 

npm install -g cubejs-cli

   

Or you can use npx without global installation:

 

npx cubejs-cli

   

4. Create a New Cube Project

To create a Cube application connected to PostgreSQL:

 

cubejs create my-cube-app -d postgres

   

Or using npx:

 

npx cubejs-cli create my-cube-app -d postgres

   

Here:

·      my-cube-app: project name

·      -d postgres: configures PostgreSQL as the database

 

Cube will automatically generate the initial project structure and configuration files.

 

Navigate to the Project

Once the project is created:

 

cd my-cube-app

Install Dependencies

Run following command.  

 

npm install

   

This installs all required project dependencies.

 

Start Cube Locally

Now start the development server:

 

npm run dev

Cube will start locally, usually at:

http://localhost:4000

 

You can able to see the user interface like below.

 

You will be redirected to the database connection setup page. Choose the database of your choice and complete the configuration.

 

For example, I configured all the PostgreSQL settings and applied the changes.

 

Once the database is configured, you will be redirected to the Cube Data Exploration page.

 

What Happens After Startup?

Once Cube is running:

·      Cube launches its local API server

·      The semantic layer becomes available

·      You can define cubes and metrics

·      SQL queries are generated automatically

·      APIs become available for analytics consumption

 

You can also access Cube Playground in the browser to:

·      Explore schemas

·      Generate queries

·      Test measures and dimensions

·      Preview generated SQL

·      Build dashboards faster

 

Previous                                                    Next                                                    Home