KV is pull based so the data is stored
KV is pull-based so the data is stored in two central locations and cached at the colo the
.get runs in for 60 seconds (default) - you can cache it for longer with .get(key, { cacheTtl: seconds }); if you wasn't already
it depends what you're doing though - some things are faster to do in the worker everytime than they might be to cache in Cache API/KV & pull from2 Replies
That is very useful info, so I could easily re-introduce KV by simply supplying a long TTL like 1 day or so and we'd be in business! I'm going to give that a try and report results back...
Though note that a high CacheTTL is not a guarantee that it will be cached that long. In fact, if it isn’t requested often enough, it may be evicted a lot earlier