On Fri, Jun 10, 2005 at 01:43:39PM -0700, Dan Stromberg wrote: > Consider: > > You have a bunch of "bricks" that can shuffle data between a NAS head > and a bunch of disks. > > The disks are RAID'd (through the "bricks"), but if one of the bricks > themselves dies, you're kinda stuck. > > But if you RAID 5 the RAID 5's, then you don't end up with massive > parity pounding, and your bricks aren't a single point of failure, and > you don't lose as much space as if you mirrored. Off the top of my head, this is what I am thinking, but I could well have missed something... Assume you have 50 disks. If you organise them as 25 two disk RAID 1s and then RAID 0 the RAID 1s you end up with the capacity of 25 disks. Any given read will benefit from a 25-way stripe; any given write suffers a 2-way mirror. If a disk dies then it alone will be rebuilt from its single mirror pair. You can lose up to 25 disks and still not lose any data, as long as no more than 1 disk from each pair survives. If you organise them as 10 five disk RAID 5s and then RAID 5 the RAID 5s you end up with the capacity of (5-1)*(10-1)=36 disks. Depending on your RAID technology, reads may be as fast as a 10-way stripe. As far as I can see though, a write would have to be striped to 10 RAID 5s, which would itself be striped to 5 disks each, so it would be a 50-way write. You could stand to lose a maximum of 1 disk in each of the low-level RAID 5s, plus one of the top-level RAID 5s, so I suppose the maximum you could get away with would be 10 (one from each low-level) plus the other 4 from a single low-level, for a total of 14. If you lose a single disk then you'll need a parity rebuild from 4 disks. If you lost a low-level RAID 5 then I'm not sure how that would work; it would be reading to rebuild from parity so presumably it would take advantage of the stripe and be like only reading from 9 disks? With 20 disks, the RAID 10 scenario ends up with 10 disks of capacity, maximum 10 disks can fail, reads are as a 10-way stripe, writes are as a 2-way mirror. The RAID 55 scenario assuming 4-disk low-level RAID 5s would be 15 disks of capacity, could theoretically have 8 disks fail and still run degraded. I think reads would be as from a 5-way stripe, with writes striped across 20 disks. So, unless I have misunderstood, depending on how you split the RAID 5s you'll get about 75% of the disk as opposed to 50% for RAID 10, but the write performance and the reliability seem much worse.
Attachment:
signature.asc
Description: Digital signature
