| 1 |
On Friday, December 8, 2017 12:48:45 AM CET Wols Lists wrote: |
| 2 |
> On 07/12/17 22:35, Frank Steinmetzger wrote: |
| 3 |
> >> (Oh - and md raid-5/6 also mix data and parity, so the same holds true |
| 4 |
> >> |
| 5 |
> >> > there.) |
| 6 |
> > |
| 7 |
> > Ok, wasn’t aware of that. I thought I read in a ZFS article that this were |
| 8 |
> > a special thing. |
| 9 |
> |
| 10 |
> Say you've got a four-drive raid-6, it'll be something like |
| 11 |
> |
| 12 |
> data1 data2 parity1 parity2 |
| 13 |
> data3 parity3 parity4 data4 |
| 14 |
> parity5 parity6 data5 data6 |
| 15 |
> |
| 16 |
> The only thing to watch out for (and zfs is likely the same) if a file |
| 17 |
> fits inside a single chunk it will be recoverable from a single drive. |
| 18 |
> And I think chunks can be anything up to 64MB. |
| 19 |
|
| 20 |
Except that ZFS doesn't have fixed on-disk-chunk-sizes. (especially if you use |
| 21 |
compression) |
| 22 |
|
| 23 |
See: |
| 24 |
https://www.delphix.com/blog/delphix-engineering/zfs-raidz-stripe-width-or-how-i-learned-stop-worrying-and-love-raidz |
| 25 |
|
| 26 |
-- |
| 27 |
Joost |
Archives