ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
ZSTD Relative Compression Performance
2
System: Intel Xeon E5-2630 v4 @ 2.20GHz (10 cores, 20 threads), 32GB RAM, 4x 500 Samsing 860 EVO SSDs
3
ZFS config: ashift=12 atime=off recordsize=128k arc_max=256m
4
Workload: fio uncompressible data to SSD: bs=128k rw=randwrite numjobs=8 iodepth=16 size=2G end_fsync=1 ioengine=posixaio runtime=30
5
Notes: fastest levels ran into the limitations of the platform (SSD i/o)
6
7
Compressionkbytes/sec
8
off1150875
9
lz41130840
10
zstd-fast-10001115378
11
zstd-fast-5001117699
12
zstd-fast-1001125854
13
zstd-fast-101112417
14
zstd-fast-11121821
15
zstd-11128019
16
zstd-21025030
17
zstd-3542716
18
zstd-4511013
19
zstd-5345419
20
zstd-6317337
21
zstd-7288826
22
zstd-8274318
23
zstd-9244768
24
zstd-10206358
25
zstd-11163317
26
zstd-12126038
27
zstd-13155655
28
zstd-1494730
29
zstd-1584605
30
zstd-1680789
31
zstd-1770569
32
zstd-1872333
33
zstd-1943638
34
gzip-1277523
35
gzip-9174498
36
37
38
39
40
41
42
43
This is a random write test to my pool made of 4 SSDs. This ran into the performance limitations of the SSDs when testing the very fast algorithms.
Since the data generated by fio is completely uncompressible, there is no gain from the higher compression levels.
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100