This looks really cool, but I'm always instinctively turned off by the pull model.

- Assuming metrics are stored in memory, they'll be lost if the app or server recycles. Of course, a push model should probably batch so there's probably some kind of overlap here.

- I've heard pull is good for detecting if a server is down. The flip side is no automatic discovery and dynamic scaling where a server going down might not necessarily be a bad thing.

- Security. Not super thrilled with exposing something that says "Hey come get my data".