Score:1

Redundancy of Raid 10 with a hot spare?

us flag

Raid 10 has a 2N redundancy. What is the redundancy for Raid 10 with a hot spare?

Am I correct in thinking it is 2N+1 redundancy?

in flag
A spare does not change anything, if one drive fails then it can rebuild, but if one more drive fails before it completes you can still be screwed.
Score:2
cn flag

Am I correct in thinking it is 2N+1 redundancy?

No, why would it? Consider thinking about edge cases.

Redundancy is 2N - data is stored on 2 discs, able to handle a failure.

What in your world does a hot spare - that takes HOURS TO INITIALIZE - actually change? Here is a hint: NOTHING. If one drive fails, there is no redundancy - UNTIL THE HOT SPARE HAS INITIALIZED. Which, again, can take hours.

I once had a Raid 6 failing while I watched it. Within like 5 minutes 3 discs went out. 2 Hot spares changed NOTHING - nada. Exactly this scenario you describe.

Your failure: assuming a hot spare does not take time to initialize.

Score:0
ru flag

Raid 10 has a 2N redundancy.

Not really. You can't have any two drives fail, just one drive from each mirror. So, the reduncancy is N+1 and after failure, one stripe has no redundancy left, the other remains at N+1.

What is the redundancy for Raid 10 with a hot spare?

Still N+1. A hot spare does not add redundancy.

A hot spare decreases the time from failure (prediction) to begin of rebuild. Without a hot spare, a failed drive requires manual replacement which may take from minutes to days. In some scenarios it can also decrease the rebuild time (when failure is predicted but the failing drive can still be copied).

If you require N+2 redundancy you'll need RAID 6(0) and should still consider hot spares.

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.