ZFS Data Deduplication

In solaris 11, you can use the deduplication (dedup) property to remove redundant data from your ZFS file systems.

To turn in on use

zfs set dedup=on rpool

But before you do that make sure that you need it

# zdb -S rpool
Simulated DDT histogram:

bucket allocated referenced
______ ______________________________ ______________________________
refcnt blocks LSIZE PSIZE DSIZE blocks LSIZE PSIZE DSIZE
—— —— —– —– —– —— —– —– —–
1 116K 8.14G 8.14G 8.14G 116K 8.14G 8.14G 8.14G
2 3.12K 101M 101M 101M 6.72K 208M 208M 208M
4 788 11.7M 11.7M 11.7M 3.73K 60.5M 60.5M 60.5M
8 260 460K 460K 460K 2.71K 5.10M 5.10M 5.10M
16 76 42K 42K 42K 1.63K 918K 918K 918K
32 30 20.5K 20.5K 20.5K 1.10K 764K 764K 764K
64 11 8K 8K 8K 991 718K 718K 718K
128 1 1.50K 1.50K 1.50K 138 207K 207K 207K
256 2 1K 1K 1K 979 490K 490K 490K
1K 2 130K 130K 130K 3.01K 176M 176M 176M
Total 120K 8.25G 8.25G 8.25G 136K 8.58G 8.58G 8.58G

dedup = 1.04, compress = 1.00, copies = 1.00, dedup * compress / copies = 1.04

Here the dedup ratio (dedup = 1.04) is less than 2, so enabling is not recommended.

Leave a Reply

Your email address will not be published. Required fields are marked *