This is a really good crash course into operating a Redshift cluster. It has a few things missing, but overall covers most the things you'll run into on a day-t

This is a really good crash course into operating a Redshift cluster. It has a few things missing, but overall covers most the things you'll run into on a day-to-day basis. After several years of operating a cluster, the two biggest headaches I ran into are WLM tuning and text-heavy data. If you're not having problems with WLM/query response time and you're not using Redshift for text-heavy datasets, cluster management is generally pretty easy and straightforward. Once your usage grows enough you actually need to pay attention to it, WLM becomes a bit of black magic voodoo dancing to find an effective WLM policy combined with a bit of utilization monitoring to know when to perform your next WLM voodoo ritual. And Redshift is incredibly handy when you need to perform SQL queries over generally-free-text data. But there are a lot of catches if you do, all the way from having to be careful of Redshift's assumptions on data you're importing to unintuitive error messages if you run up against some internal limitations during a SQL query.