Joseph Schorr
Joseph Schorr
@benjamin99 Yeah, probably, which is why we didn't do so in the first place... but maybe its fast enough that it doesn't matter? We'd have to benchmark and see
@benjamin99 Happy Lunar New Year! (apologies if that's not the correct way to say so) And no rush - I'll see if I can run some benchmarks as well
@benjamin99 Checking in
> Some early feedback, I'll continue tomorrow. Please describe in the PR body what problems are you trying to solve and design choices you took to come up with this...
> We are transferring UUIDs in every API call for no reason other than to prevent an eventual migration within the same datastore type. That's **extremely** wasteful. Is it really?...
Updated to only store 8 bytes of prefix of the datastore ID. Since datastores are *rarely* updated, this should be fine for detecting changes.
Separate clusters also allows for different preshared keys to be used, so you can have a "read only" key