postgres-sample-dbs

A collection of sample Postgres databases for learning, testing, and development.

How the dataset files were created

Data was loaded into Neon Serverless Postgres (Postgres 15). The data was then dumped using the pg_dump utility. For example:

pg_dump "postgres://<user>:<password>@<hostname>/<dbname>" --file=[file_name].sql --format=p --no-owner --no-privileges

For larger datasets, such as the employees database, the following format option was used: --format=c

Clone the repository to your local machine

git clone https://github.com/danieltprice/postgres-sample-dbs.git

Download an individual dump file

You can download an individual dump file from this repo on the GitHub site or using wget.

From this repo on the GitHub site:

1. Click on the dump file to open it.
2. Above the content of the file, you should see a button labeled "Raw". Click it. This will open a new tab or window in your browser displaying the raw contents of the file.
3. Right-click anywhere in the window or tab displaying the raw file contents, and select "Save As..." or "Save Page As ..." from the context menu. Choose a location on your computer to save the file, and click "Save".

Using wget:

get https://raw.githubusercontent.com/danieltprice/postgres-sample-dbs/main/<dump_file_name.sql>

Prerequisites

• A psql client for connecting to your Neon database and loading data. This client is included with a standalone PostgreSQL installation. See PostgreSQL Downloads.
• A pg_restore client if you are loading the employees or postgres_air database. The pg_restore client is also included with a standalone PostgreSQL installation. See PostgreSQL Downloads.
• A Neon database connection string to load data and connect to your database. After creating a database, you can obtain the connection string from the Connection Details widget on the Neon Dashboard. In the instructions that follow, replace postgres://<user>:<password>@<hostname>/[dbname] with your Neon database connection string. For further information, see Connect from any application.
• Instructions for each dataset require that you create a database. You can do so from a client such as psql or from the Neon SQL Editor.
• A Neon Pro account is required to install datasets larger than 3 GBs.

Sample data

Sample datasets are listed in order of the smallest to largest installed size. Please be aware that the Neon Free Tier has a storage limit of 3 GBs per branch. Datasets larger than 3 GBs cannot be loaded on the Free Tier.

Name	Tables	Records	Source file size	Installed size
Periodic table data	1	118	17 KB	7.2 MB
World Happiness Index	1	156	9.4 KB	7.2 MB
Titanic passenger data	1	1309	220 KB	7.5 MB
Netflix data	1	8807	3.2 MB	11 MB
Pagila database	33	62322	3 MB	15 MB
Chinook database	11	77929	1.8 MB	17 MB
Lego database	8	633250	13 MB	42 MB
Employees database	6	3919015	34 MB	333 MB
Wikipedia vector embeddings	1	25000	1.7 GB	850 MB
Postgres air	10	67228600	1.2 GB	6.7 GB

Installed size is measured using the query: `SELECT pg_size_pretty(pg_database_size('your_database_name'))`. The reported size for small datasets may appear larger than expected due to inherent Postgres storage overhead.

Periodic table data

A table containing data about the periodic table of elements.

1. Create a periodic_table database:

   CREATE DATABASE periodic_table;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/periodic_table.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/periodic_table" -f periodic_table.sql

4. Connect to the periodic_table database:

   psql postgres://<user>:<password>@<hostname>/periodic_table

5. Look up the the element with the Atomic Number 10:

   SELECT * FROM periodic_table WHERE "AtomicNumber" = 10;

• Source: https://github.com/andrejewski/periodic-table
• License: ISC License

Copyright (c) 2017, Chris Andrejewski <[email protected]>

World Happiness Index

A dataset with multiple indicators for evaluating the happiness of countries of the world.

1. Create a world_happiness database:

   CREATE DATABASE world_happiness;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/happiness_index.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/happiness_index" -f happiness_index.sql

4. Connect to the titanic database:

   psql postgres://<user>:<password>@<hostname>/world_happiness_index

5. Find the countries where the happiness score is above average but the GDP per capita is below average:

   SELECT 
       country_or_region,
       score,
       gdp_per_capita
   FROM 
       "2019"
   WHERE 
       score > (SELECT AVG(score) FROM "2019") 
       AND 
       gdp_per_capita < (SELECT AVG(gdp_per_capita) FROM "2019")
   ORDER BY 
       score DESC;

• Source: https://www.kaggle.com/datasets/unsdsn/world-happiness
• License: CC0: Public Domain

Titanic passenger data

A dataset containing information on the passengers aboard the RMS Titanic, which sank on its maiden voyage in 1912.

1. Create a titanic database:

   CREATE DATABASE titanic;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/titanic.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/titanic" -f titanic.sql

4. Connect to the titanic database:

   psql postgres://<user>:<password>@<hostname>/titanic

5. Query passengers with the most expensive fares:

   SELECT name, fare
   FROM passenger
   ORDER BY fare DESC
   LIMIT 10;

• Source: https://www.kaggle.com/datasets/ibrahimelsayed182/titanic-dataset
• License: Unknown

Netflix data

A dataset containing information about movies and tv shows on Netflix.

1. Create a netflix database:

   CREATE DATABASE netflix;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/netflix.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/netflix" -f netflix_shows.sql

4. Connect to the netflix database:

   psql postgres://<user>:<password>@<hostname>/netflix

5. Find the directors with the most movies in the database:

   SELECT 
       director,
       COUNT(*) AS "Number of Movies"
   FROM 
       netflix_shows
   WHERE 
       type = 'Movie'
   GROUP BY 
       director
   ORDER BY 
       "Number of Movies" DESC
   LIMIT 5;

• Source: https://www.kaggle.com/datasets/shivamb/netflix-shows
• License: CC0: Public Domain

Pagila database

Sample data for a fictional DVD rental store. Pagila includes tables for films, actors, film categories, stores, customers, payments, and more.

1. Create a pagila database:

   CREATE DATABASE pagila;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/pagila.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/pagila" -f pagila.sql

4. Connect to the pagila database:

   psql postgres://<user>:<password>@<hostname>/pagila

5. Find the top 10 most popular film categories based on rental frequency:

   SELECT c.name AS category_name, COUNT(r.rental_id) AS rental_count
   FROM category c
   JOIN film_category fc ON c.category_id = fc.category_id
   JOIN inventory i ON fc.film_id = i.film_id
   JOIN rental r ON i.inventory_id = r.inventory_id
   GROUP BY c.name
   ORDER BY rental_count DESC
   LIMIT 10;

• Source: https://github.com/devrimgunduz/pagila
• License: LICENSE.txt
• Copyright (c) Devrim Gündüz <[email protected]>

Chinook database

A sample database for a digital media store, including tables for artists, albums, media tracks, invoices, customers, and more.

1. Create a chinook database:

   CREATE DATABASE chinook;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/chinook.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/chinook" -f chinook.sql

4. Connect to the chinook database:

   psql postgres://<user>:<password>@<hostname>/chinook

5. Find out the most sold item by track title:

   SELECT 
   T."Name" AS "Track Title",
   SUM(IL."Quantity") AS "Total Sold"
   FROM 
       "Track" T
   JOIN 
       "InvoiceLine" IL ON T."TrackId" = IL."TrackId"
   GROUP BY 
       T."Name"
   ORDER BY 
       "Total Sold" DESC
   LIMIT 1;

• Source: https://github.com/lerocha/chinook-database
• License: LICENSE.md
• Copyright (c) 2008-2017 Luis Rocha

Lego database

A dataset containing information about various LEGO sets, their themes, parts, colors, and other associated data.

1. Create a lego database:

   CREATE DATABASE lego;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/lego.sql

3. Navigate to the directory where you downloaded the source file, and run the following command:

   psql -d "postgres://<user>:<password>@<hostname>/lego" -f lego.sql

4. Connect to the lego database:

   psql postgres://<user>:<password>@<hostname>/lego

5. Find the top 5 LEGO themes by the number of sets:

   SELECT lt.name AS theme_name, COUNT(ls.set_num) AS number_of_sets
   FROM lego_themes lt
   JOIN lego_sets ls ON lt.id = ls.theme_id
   GROUP BY lt.name
   ORDER BY number_of_sets DESC
   LIMIT 5;

• Source: https://www.kaggle.com/datasets/rtatman/lego-database
• License: CC0: Public Domain

Employees database

A dataset containing details about employees, their departments, salaries, and more.

1. Create the database and schema:

   CREATE DATABASE employees;
   \c employees
   CREATE SCHEMA employees;

2. Download the source file:

   wget https://raw.githubusercontent.com/neondatabase/postgres-sample-dbs/main/employees.sql.gz

3. Navigate to the directory where you downloaded the source file, and run the following command:

   pg_restore -d postgres://<user>:<password>@<hostname>/employees -Fc employees.sql.gz -c -v --no-owner --no-privileges

   Database objects are created in the employees schema rather than the public schema.

4. Connect to the employees database:

   psql postgres://<user>:<password>@<hostname>/employees

5. Find the top 5 departments with the highest average salary:

   SELECT d.dept_name, AVG(s.amount) AS average_salary
   FROM employees.salary s
   JOIN employees.department_employee de ON s.employee_id = de.employee_id
   JOIN employees.department d ON de.department_id = d.id
   WHERE s.to_date > CURRENT_DATE AND de.to_date > CURRENT_DATE
   GROUP BY d.dept_name
   ORDER BY average_salary DESC
   LIMIT 5;

• Source: The initial dataset was created by Fusheng Wang and Carlo Zaniolo from Siemens Corporate Research, and can be found in XML format at this location: http://timecenter.cs.aau.dk/software.htm. Designing the relational schema was undertaken by Giuseppe Maxia while Patrick Crews was responsible for transforming the data into a format compatible with MySQL. Their work can be accessed here: https://github.com/datacharmer/test_db. Subsequently, this information was adapted to a format suitable for PostgreSQL: https://github.com/h8/employees-database. The data was generated, and there are inconsistencies.
• License: This work is licensed under the Creative Commons Attribution-Share Alike 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-sa/3.0/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA.

Licensing

This repository is provided under the MIT License. However, please note that each individual database included in this repository is subject to its own license terms.

The MIT License applies to the scripts and other components that we created. We respect the rights of the original creators of the databases, and we only redistribute these databases in compliance with their respective licenses.

For each individual database, we have clearly indicated where the full text of the license can be found. If you choose to use any of these databases, you must comply with the terms specified in their respective licenses.