I am thinking about how to optimize drives and pool setup.
Assume there are two big drives in a system. What are the different pro’s and con’s on:
a) put both drives into a single pool and allocate the different shares from this single pool
b) create a pool for each drive and distribute the shares over these two pools.
Currently I created two different pools (one for each drive), and this is the reason for it:
in case one of the drive fails (maybe by a hardware defect, or by an error in the data), the failure is limited to this drive/pool, and the other one is not affected.
But maybe there are other reasons for a) or b), which I am not seeing at the moment.
I do NOT want to use RAID to duplicate everything, because this mainly duplicates the power consumption. Minimizing power consumption is essential (e.g. for environmental reasons).
As I do not require full 24/7 availability, therefore raid is a waste of energy.
Of course I have regular backups, so in case of errors I am not in trouble (but of course I want to minimize the risk for errors).
I am interested in your ideas.