Dimitris Sarlis
Dimitris Sarlis
@oggy-dfin Should we rephrase the title of this PR to something more descriptive? Like "feat: Introduce best-effort response messages" (I guess the timeout on responses is the implementation detail of...
> @mraszyk I didn't make any changes to the formal model in this PR yet, as I first wanted to get your feedback that the proposal as-is looks good. @ielashi...
> I could see some of them going into the spec, but I don't know if they all make sense given that, at least in theory, the spec should be...
Is there a specific concern that you have or is it mostly about consistency/uniformity of the APIs?
You can store vectors already. However, you would need to implement storable by hand for them. This issue is about implementing `Storable` for a `Vec` where `T` implements `Storable` so...
Hi Stan, What you describe is indeed expected, i.e. a direct write into stable memory will have less overhead than using a StableBTreeMap. We have also had other people reporting...
@ufoscout Not yet but I expect that we can look into this in the near future.
The best you can do for now would be to measure where you do these accesses (e.g. with canbench) and see if you can optimize on your side, by e.g....
> Hey folks, @dsarlis @schneiderstefan @mraszyk. I've changed this PR to completely remove the feature flag. PTAL, and we'll merge it when we feel confident... Are you sure this is...
@mraszyk I realized while working on the replica side of things that since certain parts of the canister's state, more specifically its system state, will change during replicated query execution,...