AdventureWorks for Postgres

This project provides the scripts necessary to set up the OLTP part of the go-to database used in training classes and for sample apps on the Microsoft stack. The result is 68 tables containing HR, sales, product, and purchasing data organized across 5 schemas. It represents a fictitious bicycle parts wholesaler with a hierarchy of nearly 300 employees, 500 products, 20000 customers, and 31000 sales each having an average of 4 line items. So it's big enough to be interesting, but not unwieldy. In addition to being a well-rounded OLTP sample, it is also a good choice to demonstrate ETL into a data warehouse.

Provided is a ruby file to convert CSVs available on CodePlex into a format usable by Postgres, as well as a Postgres script to create the tables, load the data, convert the hierarchyid columns, add primary and foreign keys, and create some of the views used by Adventureworks.

How to set up the database:

Download Adventure Works 2014 OLTP Script.

Extract the .zip and copy all of the CSV files into the same folder, also containing update_csvs.rb file and install.sql.

Modify the CSVs to work with Postgres by running:

ruby update_csvs.rb

Create the database and tables, import the data, and set up the views and keys with:

psql -c "CREATE DATABASE \"Adventureworks\";"
psql -d Adventureworks < install.sql

All 68 tables are properly set up, and 11 of the 20 views are established. The ones not built are those that rely on XML functions like value and ref. To see a list of tables, open psql, and then connect to the database and show all the tables with these two commands:

\c "Adventureworks"
\dt (humanresources|person|production|purchasing|sales).*

Using with Docker

You can spin up a new database using Docker with docker-compose up.

You will need to rename the Adventure Works 2014 OLTP Script archive to adventure_works_2014_OLTP_script.zip to get this to work!

Motivation

Five years ago I was pretty happy developing .NET apps for large organizations. The stack was mature, and good practices surrounding software development were very respected. The same kind of approach I appreciated from my days writing Java code was there, and the community was passionate.

Then along came Windows 8. The //build/ conference in September 2011 revealed its first beta, and even with that early peek at the new direction things were headed, it was clear that everything about the platform was a haphazard combination of the new Metro apps along with all the traditional control panel and options and API for classic code. It left a very bad taste in my mouth. Perhaps it would look pretty, but be very unusable. I couldn't see it ever being successful. Once the "red pill" registry setting vanished from the builds in mid-2012, the Windows 7 interface was then no longer available even in Server editions. I knew it was time for a change. For a year I stuck it out watching to see if there was any hope for some kind of tablet miracle out of Redmond, but I was consistently unimpressed.

In mid-2013 a friend looped me in on a new project involving Ruby on Rails, and I fervently dove in and have very much enjoyed the elegance of that ecosystem. A big part of that has been ramping up my knowledge of Postgres. What a great database engine! I figure that others departing the Microsoft camp may appreciate the same data samples they're familiar with, so I created this along with the Northwind sample. It's been useful in the classroom training folks about Rails. I expect with the heavy-handed tactics Microsoft has now used around Windows 10 that even more organizations will choose to transition away from that platform, so there will be lots of opportunity for samples like this to help people learn a new environment.

As well, with the imminent release of SQL Server 2017 for Linux, this sample could be used to evaluate performance differences between Postgres and SQL 2017. Never thought I'd see the day that MS SQL got compiled for Linux, but alas, here we are.

Let's keep coding fun.

Enjoy!