Here's a question just from a point of view to understand how the technology works.

If there is one global hotspare & multiple disk failures at the same time (t=t1), what policy or technique decides which RAID group gets the hot spare ?

Thanks,

nal