1 |
On Sat, Feb 25, 2012 at 3:52 PM, Richard Yao <ryao@×××××××××××××.edu> wrote: |
2 |
> ZFSOnLinux performance tuning is not a priority either, but there have |
3 |
> been a few patches and the performance is good. btrfs might one day |
4 |
> outperform ZFS in terms of single disk performance, assuming that it |
5 |
> does not already, but I question the usefulness of single disk |
6 |
> performance as a performance metric. |
7 |
|
8 |
Why would btrfs be inferior to ZFS on multiple disks? I can't see how |
9 |
its architecture would do any worse, and the planned features are |
10 |
superior to ZFS (which isn't to say that ZFS can't improve either). |
11 |
|
12 |
Beyond the licensing issues ZFS also does not support reshaping of |
13 |
raid-z, which is the only n+1 redundancy solution it offers. Btrfs of |
14 |
course does not yet support n+1 at all aside from some experimental |
15 |
patches floating around, but it plans to support reshaping at some |
16 |
point in time. Of course, there is no reason you couldn't implement |
17 |
reshaping for ZFS, it just hasn't happened yet. Right now the |
18 |
competition for me is with ext4+lvm+mdraid. While I really would like |
19 |
to have COW soon, I doubt I'll implement anything that doesn't support |
20 |
reshaping as mdraid+lvm does. |
21 |
|
22 |
I do realize that you can add multiple raid-zs to a zpool, but that |
23 |
isn't quite enough. If I have 4x1TB disks I'd like to be able to add |
24 |
a single 1TB disk and end up with 5TB of space. I'd rather not have |
25 |
to find 3 more 1TB hard drives to hold the data on while I redo my |
26 |
raid and then try to somehow sell them again. |
27 |
|
28 |
Rich |