I miss teaching my observability classes. I'm excited to recreate the material probably next year and if I have my way this time it's going to be "open source" so I don't have to write half a book from scratch again.

I took an information theory focused approach last time because it's what I found to be best specifically for measuring the behavior of large scale systems behavior. Statistical views have their place and utility but my experience is that they typically lead to overly noisy detection and less ability to bisect issues.

#observability