[Feature Request] Cache introspection query

[vasek]

We have a large schema (~10MB, ~250.000 lines, ~10K types). The introspection query currently takes a couple seconds to responsd. It might be useful to cache the result of introspection (the router is aware when the schema changes so it's safe to do so).

[jensneuse]

It's a good idea and we've created an internal ticket to address it. Keeping you posted here.

[StarpTech]

Hi @vasek, we're starting to look into it. Are you able to provide a reproduction? (The artifact https://cosmo-docs.wundergraph.com/cli/router/fetch would help here a lot) Did you test the router in dev mode or production? In production, we have a plan cache. The cache is disabled when you use ART, which is enabled by default in dev mode. In the Studio, we no longer fetch the introspection data from the router; we fetch it from the schema registry. Please retest it with the latest versions. Nevertheless, there is still room for improvements but before we investigate further, we would like to know if it is still an impacts you. Thank you.

[StarpTech]

I was able to reproduce with my own dataset.

[StarpTech]

Closed in favor of #605