In Redwood: perpetuum moving data between storage servers - #8 by SteavedHams @SteavedHams does say that
These decisions make sense in the context of a production FDB cluster with dedicated data volumes, as giving space back to the filesystem is unnecessary (it will be reused within Redwood) and FDB Data Distribution will move/balance key ranges around between Storage Servers which will effectively remove BTree slack and keep logical space usage roughly balanced.
But surely that’s only true when the data being moved/rebalanced sits in the same place in the BTree? If FDB detects a ‘hot range’ or similar and decides to split it into 2 shards, one of which it moves to another storage process, then it’s quite unlikely that anything else will move onto this process and take up the same ‘space’ in the tree?