RAID Level 1: Mirroring

Our first RAID level beyond striping is known as RAID level 1, or mirroring. With a mirrored system, we simply make more than one copy of each block in the system; each copy should be placed on a separate disk, of course. By doing so, we can tolerate disk failures.

In a typical mirrored system, we will assume that for each logical block, the RAID keeps two physical copies of it. Here is an example:

In the example, disk 0 and disk 1 have identical contents, and disk 2 and disk 3 do as well; the data is striped across these mirror pairs. In fact, you may have noticed that there are a number of different ways to place block copies across the disks. The arrangement above is a common one and is sometimes called RAID-10 or (RAID 1+0) because it uses mirrored pairs (RAID-1) and then stripes (RAID-0) on top of them. Another common arrangement is RAID-01 (or RAID 0+1), which contains two large striping (RAID-0) arrays, and then mirrors (RAID-1) on top of them. For now, we will just talk about mirroring assuming the above layout.

When reading a block from a mirrored array, the RAID has a choice: it can read either copy. For example, if a read to logical block 5 is issued to the RAID, it is free to read it from either disk 2 or disk 3. When writing a block, though, no such choice exists: the RAID must update both copies of the data, in order to preserve reliability. Do note, though, that these writes can take place in parallel. For example, a write to logical block 5 could proceed to disks 2 and 3 at the same time.

RAID-1 Analysis

Capacity

Let us assess RAID-1. From a capacity standpoint, RAID-1 is expensive; with the mirroring level = 2, we only obtain half of our peak useful capacity. With $N$ disks of $B$ blocks, RAID-1 useful capacity is ($N. B)/2$ ...