Moving from Core to Scale - Opportunity to tidy Datasets

As the subject suggests, I am moving from Core to Scale and taking the opportunity to rationalize my datasets.

New Scale will have (at least) two pools, one on 8 x 12Tb spinning disks and one on 2 x 1TB nvme disks (on dual M.2 card).

My current pool / dataset structure has grown over time and is a bit of a mess …

  • FreeNAS - general stuff about FreeNAS including bash scripts
  • General - stuff
  • iocage - cages (obviously won’t be moving this)
  • Manuals - d/l manuals for the stuff we own (toaster, fridge, washing machine, motherboards, computer cases, cars, etc)
  • Music
  • Pictures
  • Software
  • PlexMedia - plex movies and tv shows
  • PlexMetaData - plex meta data
  • RecordedTV
  • MacFiles - files for the household mac users
  • TimeMachine - mac timemachine backup
  • Temp - temp
  • TempDump - more temp
  • TempDVD - temp for ripping DVDs

I am thinking of heading in this direction …

  • Shares - [spinning] windows (not datasets) folders under this for manuals, music, pictures, software, etc
  • Plex Media - [spinning] Movies and TV Shows
  • Plex Meta Data - [nvme]
  • Macs - [spinning] for macs
  • Temp - [spinning] temp (dvd ripping, recorded tv, MCEBuddy transcribing, etc)
  • TrueNAS - [nvme] scripts, backups and what not

Any thoughts or comments? Do I have the split between pool location (spinning / nvme) appropriate?

I also have a spare NVMe slot on my motherboard (X11SPI-TF) that I don’t know what to do with … any suggestions?

Personally, i prefer put apps data/metadata/db on the NVME pool (1 root datasets with all >> 1 dataset for each app >> on need, folders or another datasets for config/db, depending on app volumes requirement).
Everything else imho Is quite subjective. Off course, more nested datasets you use more way to restore snapshots you have, despite create folders Is faster