Hey Michael,

Mimir is a fork from Cortex where they have improved the issues from that project so I didn't really see a need to cover as GrafanaLabs is no longer actively contributing to the project.

The reason why I chose not to adopt Mimir was due to it's maturity in production environments. The first major release: v2.0 was released in 2022 so there isn't that many resources online apart from Grafana's official documentation. This means it won't be uncommon to find issues in production and not be able to get immediate answers for resolution

I had looked a Google's Managed Prometheus offering but they were quoting £50k a month which wasn't sustainable for us. In terms of the managed service, they are running their own forks of the open source prometheus-operator project except they have a time series database (monarch) which scales tremendously well.