Verizon Digital Operates 20k Servers

Very interesting use case here. Would love to get more technical details on the scale-out process. The author mentioned two pieces:

- Turning down the data retention period in the database. Wonder how long data is kept?

- Scaling out with additional worker nodes. Curious what st2 processes got scaled out? st2actionrunner, st2rulesengine, st2api?

Any other tips/tricks?