Real-time analytics

Idea

Whenever, I eat in a fastfood restaurant I’ve always wondered what if there would be a statistics of the orders such as most popular menus, or drinks, next to the order collection dashboard. While it is hundred percent unnecessary though it would be fun idea to see what is going on while you are waiting on your food to be prepared. I am a big fun of the whoppers 🍔, and I’ve always wanted to see is that most popular choice among people?

Well it’s well definitely possible to do so, in fact, we gonna do our dashboard!

Tech-stack

We are going to use

  • postgresql to store the orders, restaurants, dishes, as well as drinks
  • mimesis to create fake dishes, drinks, restaurants, orders
  • apache kafka to create fake orders as well as consume fake orders
  • grafana to see the changes in real-time

Please check the github repo realtime-order-analytics to see how to build it on your local machine, and run for repeating the project.

The most important things that made me to choose grafana was how easy to reproduce the dashboards using json build files. Without doing anything, we will see the same dashboard since everything can be provisioned in build time, and this is a huge advantage, especially when you want to share the ready solution for people to try effortlessly!

Final results

Below you can see the end product dashboard

⚠️ This is not a production grade solution, rather an experiment project