We recently ran some ACT benchmarks on the SM863, specifically, this drive: MZ7KM1T9HAJM
First, our configuration:
- Drive: Samsung SM863 - 1.8TB
- Drive Firmware Version: GXM1003Q
- System: Dell R720 with PERC H710P Mini (FW: 21.2.0-0007), 12 core Intel(R) Xeon(R) CPU E5-2630 0 @ 2.30GHz
The drive was configured as a RAID 0 volume, with no cache (WT NORA Direct NoCachedBadBBU), as directed by the aerospike doco, and partitioned to 90% of available space (12% spare by default, so a total of 22% total spare).
3x Results (24 hours)
data is act version 3.0
trans device
%>(ms) %>(ms)
slice 1 2 4 8 16 32 64 1 2 4 8 16 32 64
----- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
1 0.42 0.10 0.08 0.03 0.00 0.00 0.00 0.41 0.09 0.07 0.03 0.00 0.00 0.00
2 0.42 0.11 0.08 0.04 0.00 0.00 0.00 0.41 0.09 0.08 0.03 0.00 0.00 0.00
3 0.42 0.10 0.08 0.03 0.00 0.00 0.00 0.41 0.09 0.07 0.03 0.00 0.00 0.00
4 0.42 0.10 0.08 0.03 0.00 0.00 0.00 0.41 0.09 0.07 0.03 0.00 0.00 0.00
5 0.42 0.10 0.08 0.03 0.00 0.00 0.00 0.41 0.09 0.07 0.03 0.00 0.00 0.00
6 0.42 0.10 0.08 0.03 0.00 0.00 0.00 0.41 0.09 0.07 0.03 0.00 0.00 0.00
7 0.42 0.10 0.08 0.03 0.00 0.00 0.00 0.40 0.09 0.07 0.03 0.00 0.00 0.00
8 0.42 0.10 0.08 0.04 0.00 0.00 0.00 0.41 0.09 0.08 0.03 0.00 0.00 0.00
9 1.23 0.16 0.08 0.03 0.00 0.00 0.00 1.21 0.14 0.08 0.03 0.00 0.00 0.00
10 1.54 0.18 0.08 0.04 0.00 0.00 0.00 1.51 0.16 0.08 0.03 0.00 0.00 0.00
11 1.44 0.17 0.08 0.03 0.00 0.00 0.00 1.42 0.15 0.08 0.03 0.00 0.00 0.00
12 1.40 0.17 0.08 0.04 0.00 0.00 0.00 1.38 0.15 0.08 0.04 0.00 0.00 0.00
13 2.25 0.22 0.08 0.04 0.00 0.00 0.00 2.22 0.21 0.08 0.04 0.00 0.00 0.00
14 2.23 0.22 0.08 0.04 0.00 0.00 0.00 2.20 0.20 0.08 0.03 0.00 0.00 0.00
15 2.12 0.21 0.08 0.04 0.00 0.00 0.00 2.09 0.20 0.08 0.04 0.00 0.00 0.00
16 2.01 0.20 0.08 0.04 0.00 0.00 0.00 1.99 0.19 0.08 0.04 0.00 0.00 0.00
17 1.93 0.20 0.08 0.04 0.00 0.00 0.00 1.90 0.18 0.08 0.04 0.00 0.00 0.00
18 1.85 0.19 0.08 0.04 0.00 0.00 0.00 1.82 0.18 0.08 0.04 0.00 0.00 0.00
19 1.76 0.18 0.08 0.04 0.00 0.00 0.00 1.73 0.17 0.08 0.03 0.00 0.00 0.00
20 1.70 0.18 0.08 0.03 0.00 0.00 0.00 1.67 0.16 0.07 0.03 0.00 0.00 0.00
21 1.63 0.17 0.08 0.04 0.00 0.00 0.00 1.60 0.16 0.08 0.04 0.00 0.00 0.00
22 1.57 0.17 0.08 0.03 0.00 0.00 0.00 1.54 0.16 0.08 0.03 0.00 0.00 0.00
23 1.52 0.17 0.08 0.04 0.00 0.00 0.00 1.49 0.15 0.08 0.04 0.00 0.00 0.00
24 1.47 0.16 0.08 0.03 0.00 0.00 0.00 1.44 0.15 0.08 0.03 0.00 0.00 0.00
----- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
avg 1.29 0.16 0.08 0.04 0.00 0.00 0.00 1.27 0.14 0.08 0.03 0.00 0.00 0.00
max 2.25 0.22 0.08 0.04 0.00 0.00 0.00 2.22 0.21 0.08 0.04 0.00 0.00 0.00
6x Results (24 hours) - summary only
avg 3.12 0.26 0.10 0.04 0.00 0.00 0.00 3.08 0.20 0.05 0.04 0.00 0.00 0.00
max 3.25 0.28 0.10 0.05 0.00 0.00 0.00 3.21 0.22 0.05 0.04 0.00 0.00 0.00
9x Results (6 hours)
data is act version 3.0
trans device
%>(ms) %>(ms)
slice 1 2 4 8 16 32 64 1 2 4 8 16 32 64
----- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
1 4.61 0.39 0.13 0.06 0.00 0.00 0.00 4.54 0.29 0.04 0.03 0.00 0.00 0.00
2 4.87 0.41 0.12 0.06 0.00 0.00 0.00 4.79 0.31 0.04 0.03 0.00 0.00 0.00
3 4.86 0.41 0.12 0.06 0.00 0.00 0.00 4.79 0.32 0.04 0.03 0.00 0.00 0.00
4 4.88 0.42 0.13 0.06 0.00 0.00 0.00 4.80 0.32 0.04 0.03 0.00 0.00 0.00
5 4.88 0.42 0.13 0.06 0.00 0.00 0.00 4.81 0.32 0.04 0.03 0.00 0.00 0.00
6 4.90 0.42 0.13 0.06 0.00 0.00 0.00 4.83 0.32 0.04 0.03 0.00 0.00 0.00
----- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
avg 4.83 0.41 0.13 0.06 0.00 0.00 0.00 4.76 0.31 0.04 0.03 0.00 0.00 0.00
max 4.90 0.42 0.13 0.06 0.00 0.00 0.00 4.83 0.32 0.04 0.03 0.00 0.00 0.00
I did test 10x, but that more than 5% of requests were over 1ms - summary only
avg 5.51 0.46 0.14 0.06 0.00 0.00 0.00 5.42 0.35 0.03 0.03 0.00 0.00 0.00
max 5.58 0.47 0.14 0.06 0.00 0.00 0.00 5.49 0.36 0.03 0.03 0.00 0.00 0.00
Tests at 10x, 12x, and 18x don’t pass, but they don’t appear to throw errors either. 24x failed almost immediately, so I am currently working out the highest point without failures. I will update once that is complete (currently working on 20x).
EDIT:20x test results below:
data is act version 3.0
trans device
%>(ms) %>(ms)
slice 1 2 4 8 16 32 64 1 2 4 8 16 32 64
----- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
1 34.84 7.37 1.50 0.77 0.05 0.00 0.00 30.06 2.95 0.03 0.02 0.00 0.00 0.00
2 35.91 7.63 1.50 0.77 0.05 0.00 0.00 30.95 3.05 0.03 0.02 0.00 0.00 0.00
3 36.01 7.70 1.54 0.80 0.05 0.00 0.00 31.02 3.07 0.03 0.02 0.00 0.00 0.00
4 35.84 7.58 1.50 0.77 0.05 0.00 0.00 30.88 3.04 0.03 0.02 0.00 0.00 0.00
5 35.91 7.64 1.52 0.78 0.05 0.00 0.00 30.95 3.05 0.03 0.02 0.00 0.00 0.00
6 35.98 7.71 1.55 0.81 0.06 0.00 0.00 31.00 3.07 0.03 0.02 0.00 0.00 0.00
----- ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
avg 35.75 7.61 1.52 0.78 0.05 0.00 0.00 30.81 3.04 0.03 0.02 0.00 0.00 0.00
max 36.01 7.71 1.55 0.81 0.06 0.00 0.00 31.02 3.07 0.03 0.02 0.00 0.00 0.00
logged and noted.