PMM requires more spare capacity than IMM, due to the creation of the temporary third copies. Because PMM cycles may be long and other elements could fail, there must be enough spare capacity in the system during PMM to handle at least one other node failure. If you plan to use PMM, account for this spare capacity at the time of deployment. If capacity issues arise when entering PMM, the system will automatically abort the process.
PMM uses both the allocated spare capacity in the system and any free capacity available, allowing it to make the best use of all unused available capacity during maintenance.
The following equation summarizes the minimum requirements:
Free + Spare - 5% of the Storage Pool >= capacity of PMM node(s)
With IMM, it is possible to ignore, or skip, the spare capacity requirements needed if the maintained node fails, but it is not possible in PMM.