Allow slicing tags sent by the cloud environments
Hi friends 👋 I'm currently looking into refactoring opportunities to allow slicing the metadata for cloud environments. More specifically https://github.com/DataDog/dd-agent/blame/master/utils/cloud_metadata.py#L98-L105. The reason behind this change is that we currently deliver a high volume of metrics and the cardinality of the tags sent goes way over >1K which slows down a bunch of our metrics.
I've notice bits of the code being moved into the MetadataCollector and I was wondering if I could do anything to help normalize how the metadata is gathered by environment. Then I would love to expose a configuration to establish a mapping between the values gathered and the one sent by the agent.
@xvello, @hkaj I've noticed you've been playing with this quite a bit lately, if there's anything I could do to help I'd love to contribute without slowing you guys down!
What do you mean by slicing it? What kind of configuration are you thinking about?
I'm not 100% sure just yet, I'll take a few hours this week and try to come up with a proof of concept, I'll ping you guys back for feedback 🙌