Potential optimisations

This is a great service and should help reduce Redis misused. I'm really excited about it.

image

A few potential changes for future iterations:

  1. Reduce the frequency. Currently the script seems to run hourly. Would it be worth considering reducing this to daily?
  2. Run the script on the host to reduce network traffic. Currently the script generates a lot of traffic. I'm not sure of the costs associated by if we could run it locally it might be better.
  3. Choose a different time from :00. Currently lots of things run on the hour. Having a signature time that is unique for a certain cron may make it easier to detected slowdowns related to the cron in future. It also spreads work out.

cc @cmiskell @rnienaber