This question already has an answer here:

What is the big difference between software RAID and hardware RAID at the low level?

What is really going on that sets the two apart?

As far as I know, the only difference is the RAID card can have cache and it has its own processor on-board. But are there other aspects that go in the background that can make one better then the other in certain situations?

marked as duplicate by Ramhound, Dave, Journeyman Geek May 18 '14 at 15:32

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


The main difference is, that with "software RAID" (quotes intended), the OS performs and manages the task of slicing, dicing, combining, striping, parity-ing, caching and whatever the individual disks (or disklike devices such as SSDs), while with "hardware RAID" these tasks are outsourced to a secondary system running one or more secondary processors (and most possibly ASICS), running their own software, mostly called firmware.

Both approaches have their respective pros and cons: E.g. while using a specialized hardware subsystem may save resources on the main system, this comes at a cost: In the event of a catastrophic failure, disks under direct OS control are much more likely to allow recovery of data.

Not the answer you're looking for? Browse other questions tagged or ask your own question.