Given your background, I'm going to take this opportunity to ask some very noobish questions. (I will be doing my due diligence to read up on Prometheus, I lik

Given your background, I'm going to take this opportunity to ask some very noobish questions. (I will be doing my due diligence to read up on Prometheus, I likely should have done so prior, but these are just some off the top of my head that I'm not sure I'd gain a good intuition for until late in the learning process) - How is the bulk of this additional compression derived? Is it explicitly the existence of a data model that lets you use more aggressive/intelligent compression strategies? - Does this come at a cost? (increased CPU overhead, latency at read time, something like that.)