On 12/26/05, David Lang <dlang@invendra.net> wrote:
> raid5 writes n+1 blocks not n+n/2 (unless n=2 for a 3-disk raid). you can
> have a 15+1 disk raid5 array for example
>
> however raid1 (and raid10) have to write 2*n blocks to disk. so if you are
> talking about pure I/O needed raid5 wins hands down. (the same 16 drives
> would be a 8+8 array)
>
> what slows down raid 5 is that to modify a block you have to read blocks
> from all your drives to re-calculate the parity. this interleaving of
> reads and writes when all you are logicly doing is writes can really hurt.
> (this is why I asked the question that got us off on this tangent, when
> doing new writes to an array you don't have to read the blocks as they are
> blank, assuming your cacheing is enough so that you can write blocksize*n
> before the system starts actually writing the data)
Not exactly true.
Let's assume you have a 4+1 RAID5 (drives A, B, C, D and E),
and you want to update drive A. Let's assume the parity
is stored in this particular write on drive E.
One way to write it is:
write A,
read A, B, C, D,
combine A+B+C+D and write it E.
(4 reads + 2 writes)
The other way to write it is:
read oldA,
read old parity oldE
write newA,
write E = oldE + (newA-oldA) -- calculate difference between new and
old A, and apply it to old parity, then write
(2 reads + 2 writes)
The more drives you have, the smarter it is to use the second approach,
unless of course A, B, C and D are available in the cache, which is the
niciest situation.
Regards,
Dawid