For several years I've been using the naming conventions to drive configuration management and other tools. Ditched rack and DC identifiers years ago because.
1. Usually not a concern.
2. When it is, it's in ipam/dcim. Plus you can usually tell from ip address.
A simple format is: app-specific-general-tierNN. foo-mysql-db-p01, foo-nginx-web-p01
Then data is nested in a hierarchy using hiera, saltstack, or custom merge behavior in cdktf. Since the names are most specific to least specific it's easy to apply configuration for each section, letting the most specific configuration override on merge.
Easiest way to automate some of the annoying bare metal/vm requirements without bringing in or writing additional tools. You rack it, set the hostname in the idrac, then let dhcp and provisioning tools do their thing.
Then data is nested in a hierarchy using hiera, saltstack, or custom merge behavior in cdktf. Since the names are most specific to least specific it's easy to apply configuration for each section, letting the most specific configuration override on merge.
Easiest way to automate some of the annoying bare metal/vm requirements without bringing in or writing additional tools. You rack it, set the hostname in the idrac, then let dhcp and provisioning tools do their thing.