The software and data we need for this workshop is available as a Docker image. This means that the only thing you need to install locally on your computer, is Docker itself.
Download and install Docker Community Edition:
Use the provided urls to download the install files:
Windows users need Hyper-V in order to run Docker. This is automatically enabled when installing Docker Desktop on Windows. If you need to enable it manually, have a look at this link
Verify the installation by running
docker version in Terminal or PowerShell
Pull the Docker image
The Docker image is available on DockerHub. It's recommended that you download it before you arrive at the workshop:
- Open PowerShell (Windows) or terminal and type in the following
docker pull oyvindbratvedt/elastic-ifi
Run the Docker image as a container
To run the Docker image as a container, open a PowerShell/terminal and type
docker run --rm -p 5601:5601 -p 9200:9200 oyvindbratvedt/elastic-ifi
This will start Elasticsearch in the Docker container, which will spend some time indexing the ~290000 traffic accidents, and then launch Kibana. Initially the console will output errors while it waits for Elasticsearch to be fully started. Depending on your computer, this could be 20 seconds to a couple of minutes.
When the accidents are finished indexing, you can open Kibana in your browser: http://localhost:5601
Set Index Pattern
You need to point Kibana to which Elasticsearch index it should fetch data from.
- In Kibana, navigate to Management and click Index Patterns.
accidentsin the Index pattern field.
- Click Next step
- In Configure settings, select timestamp in the Time Filter field name dropdown menu.
- Click Create index pattern
Get data from further back than 15 minutes
- Navigate to Discover
- In the top right corner, press Last 15 minutes
- In the Time Range menu, press Relative
- Change the From field to
50 years ago
- Click Go
- All the 289610 accidents in the dataset should now be visible
Now you can play around with the data in Kibana. Go to Visualize where you can create coordinate maps, graphs and plots visualizing the accidents.
You are now all set up to create visualisations that will help you find answers to the assignments!
Selecting the correct way to display your data can be quite challenging. Check out this link for some hints on graphs suitable for visualising many different data relationships.