In memory replication by host #7276
Description
Is your feature request related to a problem? Please describe.
Currently all replications tasks have to be hydrated before they can be replicated, creating unnecessary database requests and slowing down the replication task fetching process
Proposed Solution
Implement a bytesize cache per shard and have a budget management per host to cache replication tasks in memory, accounting for tasks imbalance per shard
Additional context
Add any other context or screenshots about the feature request here.