A hot spare is a standby drive that automatically takes over when a member fails, so the rebuild starts immediately instead of waiting for a human. Size arrays with spares in the RAID calculator.
Hot, cold and global spares
A hot spare is powered-on and idle in the array, ready to be brought in automatically the moment a drive fails — the controller starts rebuilding onto it without intervention. A cold spare sits on a shelf and must be fitted by hand. A global (or shared) hot spare can cover several arrays rather than being dedicated to one.
The value of a hot spare is time: it shrinks the dangerous window between a failure and a healthy array back to just the rebuild time, rather than rebuild time plus however long it takes someone to notice and swap a disk.
Distributed spares (RAID 5E/5EE, ADAPT, RAIDZ)
Instead of one idle spare drive, modern arrays often distribute the spare capacity across all drives. RAID 5E/5EE do this; so do Dell's ADAPT and HPE's distributed RAID, and ZFS uses spare/free space similarly. Every spindle stays active (a little more performance) and rebuilds are faster because they write across many drives at once rather than funnelling into one.
Our RAID 5E calculator shows how an integrated spare costs one drive of usable capacity in exchange for instant, self-healing failover.
Do you still need backups?
Yes. A hot spare speeds recovery from a drive failure, but it does nothing for deletion, corruption or ransomware — that's backup territory. Think of a hot spare as reducing downtime risk, not as a substitute for a second copy.
On large drives, pair a hot spare with dual parity (RAID 6) so the array survives a second failure during the rebuild.