Unraid may be the easiest way to manage these Chia plots and drives. Thank you again for this docker-app.
I ended up setting up a server with a trial license:
- Did not use parity for optimal write speeds. Parity is also not needed as lost data can be replotted. Since UnRAID has data stored at each disk, no need to worry about losing the entire array.
- I used 4x 400GB S3700s as a cache pool to catch all the plots coming from my plotters. The mover-tuning script allowed me to ignore .tmp files when the mover runs every 2hours or so. This is to prevent from causing a traffic jam at spinner-disk array level.
- Saturate my 10gbps connections.
If things get serious (more than 28 drives?), I'll start having to figure out how to use the unassigned drives as a place to store plots.
