Heyang Zhou

The design is to form a mesh network of all mvstore instances so that they can synchronize cache invalidation logs directly and keep everything in-memory. Metadata is still backed by...

@fire I haven't been able to get back to this yet... The idea I have in mind is that, we select one mvstore process in the cluster to be the...

The reason why we don't currently provide a mechanism to get whether a `delete()` actually operated on an existing key is that a blind delete with a non-discarded result is...

Yes the standard way works. Providing an API to dump on the page layer directly is more efficient though.

If you really need to cache column names for whatever reason (performance?), maybe something like this can be done: ```sql BEGIN; PRAGMA schema_version; -- compare the returned value with the...

Triggers do not cross the process boundary. SQLite internally has knowledge of the schema version, but that doesn't seem to be available to outside like column names are. A new...

Currently there isn't a cross-process notification mechanism for KV in CLI. For queues, every process periodically queries the SQLite database for updates. That's fine because there's only one queue namespace...

Latest `denokv` does support watch. Did you run into any issues? > but sqlite supports triggers which should be able resolve the polling issue. SQLite triggers work on the database...

> That being said, sqlite offers [Data Change Notifications](https://www.sqlite.org/c3ref/update_hook.html), which only set when there is something to watch would be a minimal performance impact. As [discussed on the SQLite forum](https://sqlite.org/forum/info/524b9656ce3bb216813a775e0767aa6a98c006247b37fd219643cb64fc73d13c),...

Hi, Please check [the docs](https://docs.deno.com/kv/manual/backup) for setting up sync to S3.