DigitalOcean Benchmarks DigitalOcean US$15/month droplet benchmarks - Intel Xeon E5-2630Lv2 vs Intel Xeon Gold 6140

eva2000 · Dec 17, 2018

DigitalOcean KVM VPS droplets run on a variety of cpu models. To date most of my droplets for production live sites has been SFO1 datacenter which had Intel Xeon E5-2630Lv2 IvyBridge based cpus. DigitalOcean's higher priced CPU optimised droplets had newer Intel Xeon Platinum Scalable Skylake based cpus and some folks have reported being lucky enough to get these newer Intel Xeon Scalable Skylake cpus as well.

Today I decided to create a snapshot of my existing production live site's SFO1 2 CPU, 60GB US$15/month droplet running Centmin Mod 123.09beta01 LEMP stack and move it to SFO2 region in hopes of getting a newer cpu model. I got lucky on the first attempt and managed to get 2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz

Both droplets are running

Centmin Mod 123.09beta01 LEMP stack
Nginx 1.15.7
PHP-FPM 7.2.13
MariaDB 10.0.37 MySQL
CentOS 7.6 64bit
US$15/month droplets

To compare.

DigitalOcean SFO2 Droplet - 2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz

Code (Text):

lscpu
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                2
On-line CPU(s) list:   0,1
Thread(s) per core:    1
Core(s) per socket:    1
Socket(s):             2
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 85
Model name:            Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz
Stepping:              4
CPU MHz:               2294.608
BogoMIPS:              4589.21
Virtualization:        VT-x
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              1024K
L3 cache:              25344K
NUMA node0 CPU(s):     0,1
Flags:                 fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl eagerfpu pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 pku ospke

compared to DigitalOcean SFO1 Droplet - 2 CPU, 60GB US$15/month on older Intel Xeon E5-2630L v2 at 2.40GHz

Code (Text):

lscpu
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                2
On-line CPU(s) list:   0,1
Thread(s) per core:    1
Core(s) per socket:    1
Socket(s):             2
NUMA node(s):          1
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 62
Model name:            Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz
Stepping:              4
CPU MHz:               2399.998
BogoMIPS:              4799.99
Virtualization:        VT-x
Hypervisor vendor:     KVM
Virtualization type:   full
L1d cache:             32K
L1i cache:             32K
L2 cache:              256K
L3 cache:              15360K
NUMA node0 CPU(s):     0,1
Flags:                 fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl eagerfpu pni pclmulqdq vmx ssse3 cx16 pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust smep erms xsaveopt

sysbench benchmarks

I installed my sysbench script on both servers to do some comparison tests. Results below show the newer cpu droplet with Intel Xeon Gold 6140 performing much better for cpu and memory benchmarks but both MySQL OLTP and disk I/O performance was dramatically slower on the newer cpu droplet compared to the old droplet as you will see in sysbench file, file-fsync and OLTP benchmark tests. The new cpu droplet had file fsyncs that were 52x times slower - not good for MySQL database performance and it does show on the sysbench MySQL OLTP benchmark numbers.

sysbench.sh install

Code (Text):

mkdir -p /root/tools/sysbench
cd /root/tools/sysbench
wget -O /root/tools/sysbench/sysbench.sh https://github.com/centminmod/centminmod-sysbench/raw/master/sysbench.sh
chmod +x sysbench.sh
./sysbench.sh install

sysbench cpu benchmarks show newer Intel Xeon Gold 6140 droplet up to 49% faster than the older Intel Xeon E5-2630Lv2 droplet

sysbench memory benchmarks show newer Intel Xeon Gold 6140 droplet between 22-26% faster than the older Intel Xeon E5-2630Lv2 droplet

sysbench file disk I/O sequential benchmarks show newer Intel Xeon Gold 6140 droplet

for reads/s and writes/s, up to 25% slower for 1 thread sequential read and up to 63.7% slower for 2 thread sequential read and up 86% slower for 1 thread sequential write and up to 86% slower for 2 thread sequential write compared to the older Intel Xeon E5-2630Lv2 droplet
for MB/s read and writes, up to 24% slower for 1 thread sequential read and up to 63.7% slower for 2 thread sequential read and up 86% slower for 1 thread sequential write and up to 86% slower for 2 thread sequential write compared to the older Intel Xeon E5-2630Lv2 droplet

sysbench file disk I/O random benchmarks show newer Intel Xeon Gold 6140 droplet

for reads/s and writes/s, up to 32% slower for 1 thread random read and up to 34% slower for 2 thread random read and up 97% slower for 1 thread random write and up to 93.5% slower for 2 thread random write compared to the older Intel Xeon E5-2630Lv2 droplet
for MB/s read and writes, up to 32% slower for 1 thread random read and up to 34% slower for 2 thread random read and up 97% slower for 1 thread random write and up to 93.5% slower for 2 thread random write compared to the older Intel Xeon E5-2630Lv2 droplet

sysbench file disk fsync benchmarks show newer Intel Xeon Gold 6140 droplet up to 97% slower than the older Intel Xeon E5-2630Lv2 droplet for fsync per second and 99% percentile fsync response time was 52x times slower on the newer cpu !

sysbench MySLQ OLTP benchmarks show newer Intel Xeon Gold 6140 droplet up to 48% slower transactions/s and queries/s than the older Intel Xeon E5-2630Lv2 droplet and 99% percentile MySQL response time was 2.79x times slower on the newer cpu !

Raw numbers below

2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz
cpu

Code (Text):

./sysbench.sh cpu

sysbench cpu --cpu-max-prime=20000 --threads=1 run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
prime: 20000
events/s: 388.85
time: 10.0022s
min: 2.42
avg: 2.57
max: 10.82
95th: 2.76

| cpu sysbench | threads: | events/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 1.0.16 | 1 | 388.85 | 10.0022s | 2.42 | 2.57 | 10.82 | 2.76 |

sysbench,threads,events/s,time,min,avg,max,95th
1.0.16,1,388.85,10.0022s,2.42,2.57,10.82,2.76

sysbench cpu --cpu-max-prime=20000 --threads=2 run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
prime: 20000
events/s: 769.71
time: 10.0020s
min: 2.46
avg: 2.60
max: 7.68
95th: 2.81

| cpu sysbench | threads: | events/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 1.0.16 | 2 | 769.71 | 10.0020s | 2.46 | 2.60 | 7.68 | 2.81 |

sysbench,threads,events/s,time,min,avg,max,95th
1.0.16,2,769.71,10.0020s,2.46,2.60,7.68,2.81

2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz
memory

Code (Text):

./sysbench.sh mem

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
block-size: 1KiB
total-size: 1024MiB
operation: read
scope: global
total-ops: 1048576 (5011549.00 per second)
transferred (4894.09 MiB/sec)
time: 0.2040s
min: 0.00
avg: 0.00
max: 0.04
95th: 0.00

| memory sysbench | sysbench | threads: | block-size: | total-size: | operation: | total-ops: | transferred | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| memory | 1.0.16 | 1 | 1KiB | 1024MiB | read | 1048576 | 4894.09 | 0.2040s | 0.00 | 0.00 | 0.04 | 0.00 |

sysbench,sysbench,threads,block-size,total-size,operation,total-ops,transferred,time,min,avg,max,95th
memory,1.0.16,1,1KiB,1024MiB,read,1048576,4894.09,0.2040s,0.00,0.00,0.04,0.00

sysbench memory --threads=2 --memory-block-size=1K --memory-scope=global --memory-total-size=1G --memory-oper=read run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
block-size: 1KiB
total-size: 1024MiB
operation: read
scope: global
total-ops: 1048576 (8787232.73 per second)
transferred (8581.28 MiB/sec)
time: 0.1176s
min: 0.00
avg: 0.00
max: 0.19
95th: 0.00

| memory sysbench | sysbench | threads: | block-size: | total-size: | operation: | total-ops: | transferred | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| memory | 1.0.16 | 2 | 1KiB | 1024MiB | read | 1048576 | 8581.28 | 0.1176s | 0.00 | 0.00 | 0.19 | 0.00 |

sysbench,sysbench,threads,block-size,total-size,operation,total-ops,transferred,time,min,avg,max,95th
memory,1.0.16,2,1KiB,1024MiB,read,1048576,8581.28,0.1176s,0.00,0.00,0.19,0.00

2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz
file

Code (Text):

./sysbench.sh file

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqrd prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqrd-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Using synchronous I/O mode
Doing sequential read test
reads/s: 6173.39
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 24.11
written-MiB/s: 0.00
time: 10.0003s
min: 0.06
avg: 0.16
max: 10.85
95th: 0.37

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | read | 6173.39 | 0.00 | 0.00 | 24.11 | 0.00 | 10.0003s | 0.06 | 0.16 | 10.85 | 0.37 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,read,6173.39,0.00,0.00,24.11,0.00,10.0003s,0.06,0.16,10.85,0.37


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqwr prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqwr-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Using synchronous I/O mode
Doing sequential write (creation) test
reads/s: 0.00
writes/s: 385.87
fsyncs/s: 498.39
read-MiB/s: 0.00
written-MiB/s: 1.51
time: 10.0119s
min: 0.00
avg: 1.14
max: 745.55
95th: 3.30

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 385.87 | 498.39 | 0.00 | 1.51 | 10.0119s | 0.00 | 1.14 | 745.55 | 3.30 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,write,0.00,385.87,498.39,0.00,1.51,10.0119s,0.00,1.14,745.55,3.30


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndrd prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndrd-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random read test
reads/s: 2664.98
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 10.41
written-MiB/s: 0.00
time: 10.0002s
min: 0.12
avg: 0.37
max: 11.05
95th: 0.84

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | read | 2664.98 | 0.00 | 0.00 | 10.41 | 0.00 | 10.0002s | 0.12 | 0.37 | 11.05 | 0.84 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,read,2664.98,0.00,0.00,10.41,0.00,10.0002s,0.12,0.37,11.05,0.84


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndwr prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndwr-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random write test
reads/s: 0.00
writes/s: 140.35
fsyncs/s: 191.80
read-MiB/s: 0.00
written-MiB/s: 0.55
time: 10.0055s
min: 0.00
avg: 3.13
max: 1002.51
95th: 8.13

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 140.35 | 191.80 | 0.00 | 0.55 | 10.0055s | 0.00 | 3.13 | 1002.51 | 8.13 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,write,0.00,140.35,191.80,0.00,0.55,10.0055s,0.00,3.13,1002.51,8.13



sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqrd prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqrd-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Using synchronous I/O mode
Doing sequential read test
reads/s: 5347.19
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 20.89
written-MiB/s: 0.00
time: 10.0004s
min: 0.06
avg: 0.37
max: 17.16
95th: 1.01

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | read | 5347.19 | 0.00 | 0.00 | 20.89 | 0.00 | 10.0004s | 0.06 | 0.37 | 17.16 | 1.01 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,read,5347.19,0.00,0.00,20.89,0.00,10.0004s,0.06,0.37,17.16,1.01


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqwr prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqwr-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Using synchronous I/O mode
Doing sequential write (creation) test
reads/s: 0.00
writes/s: 444.72
fsyncs/s: 587.90
read-MiB/s: 0.00
written-MiB/s: 1.74
time: 10.0105s
min: 0.00
avg: 1.98
max: 223.48
95th: 6.43

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 444.72 | 587.90 | 0.00 | 1.74 | 10.0105s | 0.00 | 1.98 | 223.48 | 6.43 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,write,0.00,444.72,587.90,0.00,1.74,10.0105s,0.00,1.98,223.48,6.43


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndrd prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndrd-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random read test
reads/s: 5276.96
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 20.61
written-MiB/s: 0.00
time: 10.0002s
min: 0.12
avg: 0.38
max: 10.19
95th: 0.86

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | read | 5276.96 | 0.00 | 0.00 | 20.61 | 0.00 | 10.0002s | 0.12 | 0.38 | 10.19 | 0.86 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,read,5276.96,0.00,0.00,20.61,0.00,10.0002s,0.12,0.38,10.19,0.86


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndwr prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndwr-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random write test
reads/s: 0.00
writes/s: 480.28
fsyncs/s: 638.65
read-MiB/s: 0.00
written-MiB/s: 1.88
time: 10.0182s
min: 0.00
avg: 1.82
max: 34.69
95th: 6.79

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 480.28 | 638.65 | 0.00 | 1.88 | 10.0182s | 0.00 | 1.82 | 34.69 | 6.79 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,write,0.00,480.28,638.65,0.00,1.88,10.0182s,0.00,1.82,34.69,6.79


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|fileio | 1.0.16 | 1 | 4KiB | I/O | read | 6173.39 | 0.00 | 0.00 | 24.11 | 0.00 | 10.0003s | 0.06 | 0.16 | 10.85 | 0.37 |
|fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 385.87 | 498.39 | 0.00 | 1.51 | 10.0119s | 0.00 | 1.14 | 745.55 | 3.30 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | read | 5347.19 | 0.00 | 0.00 | 20.89 | 0.00 | 10.0004s | 0.06 | 0.37 | 17.16 | 1.01 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 444.72 | 587.90 | 0.00 | 1.74 | 10.0105s | 0.00 | 1.98 | 223.48 | 6.43 |

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|fileio | 1.0.16 | 1 | 4KiB | I/O | read | 2664.98 | 0.00 | 0.00 | 10.41 | 0.00 | 10.0002s | 0.12 | 0.37 | 11.05 | 0.84 |
|fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 140.35 | 191.80 | 0.00 | 0.55 | 10.0055s | 0.00 | 3.13 | 1002.51 | 8.13 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | read | 5276.96 | 0.00 | 0.00 | 20.61 | 0.00 | 10.0002s | 0.12 | 0.38 | 10.19 | 0.86 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 480.28 | 638.65 | 0.00 | 1.88 | 10.0182s | 0.00 | 1.82 | 34.69 | 6.79 |

2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz
file-fsync

Code (Text):

./sysbench.sh file-fsync

sysbench fileio fsync prepare
sysbench fileio --time=30 --file-num=1 --file-extra-flags= --file-total-size=4096 --file-block-size=4096 --file-fsync-all=on --file-test-mode=rndwr --file-fsync-freq=0 --file-fsync-end=0 --threads=1 --percentile=99 prepare

sysbench fileio --threads=1 --time=30 --file-num=1 --file-extra-flags= --file-total-size=4096 --file-block-size=4096 --file-fsync-all=on --file-test-mode=rndwr --file-fsync-freq=0 --file-fsync-end=0 --percentile=99 run
raw log saved: /home/sysbench/sysbench-fileio-fsync-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random write test
reads/s: 0.00
writes/s: 95.24
fsyncs/s: 95.24
read-MiB/s: 0.00
written-MiB/s: 0.37
time: 30.0052s
min: 1.60
avg: 10.49
max: 2000.03
99th: 40.37

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 99th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 95.24 | 95.24 | 0.00 | 0.37 | 30.0052s | 1.60 | 10.49 | 2000.03 | 40.37 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,99th
fileio,1.0.16,1,4KiB,I/O,write,0.00,95.24,95.24,0.00,0.37,30.0052s,1.60,10.49,2000.03,40.37


sysbench fileio cleanup
sysbench fileio --file-num=1 --file-extra-flags= --file-total-size=4096 --file-block-size=4096 cleanup

2 CPU, 60GB US$15/month on Intel Xeon Gold 6140 at 2.30Ghz
mysql oltp

Code (Text):

./sysbench.sh mysqloltpnew

MySQL Buffers
aria_pagecache_buffer_size           16777216
aria_sort_buffer_size                16777216
bulk_insert_buffer_size              8388608
innodb_buffer_pool_dump_at_shutdown  ON
innodb_buffer_pool_dump_now          OFF
innodb_buffer_pool_dump_pct          100
innodb_buffer_pool_filename          ib_buffer_pool
innodb_buffer_pool_instances         8
innodb_buffer_pool_load_abort        OFF
innodb_buffer_pool_load_at_startup   ON
innodb_buffer_pool_load_now          OFF
innodb_buffer_pool_populate          OFF
innodb_buffer_pool_size              268435456
innodb_change_buffer_max_size        25
innodb_change_buffering              all
innodb_log_buffer_size               8388608
innodb_sort_buffer_size              2097152
join_buffer_size                     262144
join_buffer_space_limit              2097152
key_buffer_size                      33554432
mrr_buffer_size                      262144
myisam_sort_buffer_size              134217728
net_buffer_length                    16384
optimizer_switch                     index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=off
preload_buffer_size                  32768
read_buffer_size                     262144
read_rnd_buffer_size                 262144
sort_buffer_size                     262144
sql_buffer_result                    OFF

MySQL Limits
aria_pagecache_division_limit         100
delayed_insert_limit                  100
expensive_subquery_limit              100
ft_query_expansion_limit              20
innodb_ft_result_cache_limit          2000000000
join_buffer_space_limit               2097152
key_cache_division_limit              100
log_slow_rate_limit                   1
min_examined_row_limit                0
open_files_limit                      524288
optimizer_selectivity_sampling_limit  100
query_cache_limit                     0
relay_log_space_limit                 0
sql_select_limit                      18446744073709551615
thread_pool_stall_limit               500
updatable_views_with_limit            YES

MySQL Maxes
aria_max_sort_file_size           9223372036853727232
extra_max_connections             2
ft_max_word_len                   84
group_concat_max_len              1024
innodb_adaptive_max_sleep_delay   150000
innodb_compression_pad_pct_max    50
innodb_file_format_max            Barracuda
innodb_ft_max_token_size          84
innodb_io_capacity_max            4000
innodb_max_bitmap_file_size       104857600
innodb_max_changed_pages          1000000
innodb_max_dirty_pages_pct        75.000000
innodb_max_dirty_pages_pct_lwm    0.001000
innodb_max_purge_lag              0
innodb_max_purge_lag_delay        0
innodb_online_alter_log_max_size  134217728
max_allowed_packet                33554432
max_binlog_cache_size             18446744073709547520
max_binlog_size                   1073741824
max_binlog_stmt_cache_size        18446744073709547520
max_connect_errors                100000
max_connections                   350
max_delayed_threads               20
max_digest_length                 1024
max_error_count                   64
max_heap_table_size               67108864
max_insert_delayed_threads        20
max_join_size                     18446744073709551615
max_length_for_sort_data          1024
max_long_data_size                33554432
max_prepared_stmt_count           16382
max_relay_log_size                1073741824
max_seeks_for_key                 1000
max_session_mem_used              9223372036854775807
max_sort_length                   1024
max_sp_recursion_depth            0
max_statement_time                0.000000
max_tmp_tables                    32
max_user_connections              0
max_write_lock_count              4294967295
myisam_max_sort_file_size         2147483648
slave_max_allowed_packet          1073741824
slave_parallel_max_queued         131072
thread_pool_max_threads           1000
wsrep_max_ws_rows                 0
wsrep_max_ws_size                 2147483647

MySQL Concurrency
concurrent_insert           ALWAYS
innodb_commit_concurrency   0
innodb_concurrency_tickets  5000
innodb_thread_concurrency   0
thread_concurrency          10

MySQL Read/Write
delay_key_write           ON
innodb_disallow_writes    OFF
innodb_doublewrite        ON
innodb_use_atomic_writes  OFF
max_write_lock_count      4294967295
net_write_timeout         60
wsrep_causal_reads        OFF
wsrep_dirty_reads         OFF

MySQL Threads
aria_repair_threads                      1
binlog_optimize_thread_scheduling        ON
debug_no_thread_alarm                    OFF
innodb_encryption_threads                0
innodb_mtflush_threads                   8
innodb_purge_threads                     1
innodb_read_io_threads                   4
innodb_thread_concurrency                0
innodb_thread_sleep_delay                10000
innodb_write_io_threads                  2
max_delayed_threads                      20
max_insert_delayed_threads               20
myisam_repair_threads                    1
performance_schema_max_thread_classes    50
performance_schema_max_thread_instances  -1
slave_domain_parallel_threads            0
slave_parallel_threads                   0
thread_cache_size                        64
thread_concurrency                       10
thread_handling                          pool-of-threads
thread_pool_idle_timeout                 60
thread_pool_max_threads                  1000
thread_pool_oversubscribe                3
thread_pool_size                         500
thread_pool_stall_limit                  500
thread_stack                             295936
wsrep_slave_threads                      1

MySQL Binlog
binlog_annotate_row_events               OFF
binlog_cache_size                        32768
binlog_checksum                          NONE
binlog_commit_wait_count                 0
binlog_commit_wait_usec                  100000
binlog_direct_non_transactional_updates  OFF
binlog_format                            STATEMENT
binlog_optimize_thread_scheduling        ON
binlog_row_image                         FULL
binlog_stmt_cache_size                   32768
encrypt_binlog                           OFF
gtid_binlog_pos
gtid_binlog_state
innodb_api_enable_binlog                 OFF
innodb_locks_unsafe_for_binlog           OFF
log_bin                                  OFF
log_bin_basename
log_bin_index
log_bin_trust_function_creators          OFF
max_binlog_cache_size                    18446744073709547520
max_binlog_size                          1073741824
max_binlog_stmt_cache_size               18446744073709547520
sql_log_bin                              ON
sync_binlog                              0
wsrep_forced_binlog_format               NONE

MySQL InnoDB
ignore_builtin_innodb                        OFF
innodb_adaptive_flushing                     ON
innodb_adaptive_flushing_lwm                 10.000000
innodb_adaptive_hash_index                   ON
innodb_adaptive_hash_index_partitions        1
innodb_adaptive_max_sleep_delay              150000
innodb_additional_mem_pool_size              8388608
innodb_api_bk_commit_interval                5
innodb_api_disable_rowlock                   OFF
innodb_api_enable_binlog                     OFF
innodb_api_enable_mdl                        OFF
innodb_api_trx_level                         0
innodb_autoextend_increment                  64
innodb_autoinc_lock_mode                     1
innodb_background_scrub_data_check_interval  3600
innodb_background_scrub_data_compressed      OFF
innodb_background_scrub_data_interval        604800
innodb_background_scrub_data_uncompressed    OFF
innodb_buf_dump_status_frequency             0
innodb_buffer_pool_dump_at_shutdown          ON
innodb_buffer_pool_dump_now                  OFF
innodb_buffer_pool_dump_pct                  100
innodb_buffer_pool_filename                  ib_buffer_pool
innodb_buffer_pool_instances                 8
innodb_buffer_pool_load_abort                OFF
innodb_buffer_pool_load_at_startup           ON
innodb_buffer_pool_load_now                  OFF
innodb_buffer_pool_populate                  OFF
innodb_buffer_pool_size                      268435456
innodb_change_buffer_max_size                25
innodb_change_buffering                      all
innodb_checksum_algorithm                    INNODB
innodb_checksums                             ON
innodb_cleaner_lsn_age_factor                HIGH_CHECKPOINT
innodb_cmp_per_index_enabled                 OFF
innodb_commit_concurrency                    0
innodb_compression_algorithm                 zlib
innodb_compression_failure_threshold_pct     5
innodb_compression_level                     6
innodb_compression_pad_pct_max               50
innodb_concurrency_tickets                   5000
innodb_corrupt_table_action                  assert
innodb_data_file_path                        ibdata1:10M:autoextend
innodb_data_home_dir
innodb_default_encryption_key_id             1
innodb_default_row_format                    dynamic
innodb_defragment                            ON
innodb_defragment_fill_factor                0.900000
innodb_defragment_fill_factor_n_recs         20
innodb_defragment_frequency                  40
innodb_defragment_n_pages                    7
innodb_defragment_stats_accuracy             0
innodb_disable_sort_file_cache               OFF
innodb_disallow_writes                       OFF
innodb_doublewrite                           ON
innodb_empty_free_list_algorithm             BACKOFF
innodb_encrypt_log                           OFF
innodb_encrypt_tables                        OFF
innodb_encryption_rotate_key_age             1
innodb_encryption_rotation_iops              100
innodb_encryption_threads                    0
innodb_fake_changes                          OFF
innodb_fast_shutdown                         1
innodb_fatal_semaphore_wait_threshold        600
innodb_file_format                           Barracuda
innodb_file_format_check                     ON
innodb_file_format_max                       Barracuda
innodb_file_per_table                        ON
innodb_flush_log_at_timeout                  1
innodb_flush_log_at_trx_commit               2
innodb_flush_method                          O_DIRECT
innodb_flush_neighbors                       1
innodb_flushing_avg_loops                    30
innodb_force_load_corrupted                  OFF
innodb_force_primary_key                     OFF
innodb_force_recovery                        0
innodb_foreground_preflush                   EXPONENTIAL_BACKOFF
innodb_ft_aux_table
innodb_ft_cache_size                         8000000
innodb_ft_enable_diag_print                  OFF
innodb_ft_enable_stopword                    ON
innodb_ft_max_token_size                     84
innodb_ft_min_token_size                     3
innodb_ft_num_word_optimize                  2000
innodb_ft_result_cache_limit                 2000000000
innodb_ft_server_stopword_table
innodb_ft_sort_pll_degree                    2
innodb_ft_total_cache_size                   640000000
innodb_ft_user_stopword_table
innodb_idle_flush_pct                        100
innodb_immediate_scrub_data_uncompressed     OFF
innodb_instrument_semaphores                 OFF
innodb_io_capacity                           2000
innodb_io_capacity_max                       4000
innodb_kill_idle_transaction                 0
innodb_large_prefix                          OFF
innodb_lock_schedule_algorithm               fcfs
innodb_lock_wait_timeout                     50
innodb_locking_fake_changes                  ON
innodb_locks_unsafe_for_binlog               OFF
innodb_log_arch_dir                          ./
innodb_log_arch_expire_sec                   0
innodb_log_archive                           OFF
innodb_log_block_size                        512
innodb_log_buffer_size                       8388608
innodb_log_checksum_algorithm                INNODB
innodb_log_compressed_pages                  ON
innodb_log_file_size                         134217728
innodb_log_files_in_group                    2
innodb_log_group_home_dir                    ./
innodb_lru_scan_depth                        1024
innodb_max_bitmap_file_size                  104857600
innodb_max_changed_pages                     1000000
innodb_max_dirty_pages_pct                   75.000000
innodb_max_dirty_pages_pct_lwm               0.001000
innodb_max_purge_lag                         0
innodb_max_purge_lag_delay                   0
innodb_mirrored_log_groups                   1
innodb_monitor_disable
innodb_monitor_enable
innodb_monitor_reset
innodb_monitor_reset_all
innodb_mtflush_threads                       8
innodb_old_blocks_pct                        37
innodb_old_blocks_time                       1000
innodb_online_alter_log_max_size             134217728
innodb_open_files                            1000
innodb_optimize_fulltext_only                OFF
innodb_page_size                             16384
innodb_prefix_index_cluster_optimization     OFF
innodb_print_all_deadlocks                   OFF
innodb_print_lock_wait_timeout_info          OFF
innodb_purge_batch_size                      300
innodb_purge_threads                         1
innodb_random_read_ahead                     OFF
innodb_read_ahead_threshold                  56
innodb_read_io_threads                       4
innodb_read_only                             OFF
innodb_replication_delay                     0
innodb_rollback_on_timeout                   OFF
innodb_rollback_segments                     128
innodb_sched_priority_cleaner                19
innodb_scrub_log                             OFF
innodb_scrub_log_speed                       256
innodb_show_locks_held                       10
innodb_show_verbose_locks                    0
innodb_simulate_comp_failures                0
innodb_sort_buffer_size                      2097152
innodb_spin_wait_delay                       6
innodb_stats_auto_recalc                     ON
innodb_stats_include_delete_marked           OFF
innodb_stats_method                          nulls_equal
innodb_stats_modified_counter                0
innodb_stats_on_metadata                     OFF
innodb_stats_persistent                      ON
innodb_stats_persistent_sample_pages         20
innodb_stats_sample_pages                    8
innodb_stats_traditional                     ON
innodb_stats_transient_sample_pages          8
innodb_status_output                         OFF
innodb_status_output_locks                   OFF
innodb_strict_mode                           OFF
innodb_support_xa                            ON
innodb_sync_array_size                       1
innodb_sync_spin_loops                       30
innodb_table_locks                           ON
innodb_thread_concurrency                    0
innodb_thread_sleep_delay                    10000
innodb_tmpdir
innodb_track_changed_pages                   OFF
innodb_undo_directory                        .
innodb_undo_logs                             128
innodb_undo_tablespaces                      0
innodb_use_atomic_writes                     OFF
innodb_use_fallocate                         OFF
innodb_use_global_flush_log_at_trx_commit    ON
innodb_use_mtflush                           OFF
innodb_use_native_aio                        ON
innodb_use_stacktrace                        OFF
innodb_use_sys_malloc                        ON
innodb_use_trim                              OFF
innodb_version                               5.6.41-84.1
innodb_write_io_threads                      2


sysbench prepare database: sbt
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql prepare
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)

Initializing worker threads...

Creating table 'sbtest2'...
Creating table 'sbtest1'...
Inserting 150000 records into 'sbtest1'
Inserting 150000 records into 'sbtest2'
Creating a secondary index on 'sbtest1'...
Creating a secondary index on 'sbtest2'...
Creating table 'sbtest3'...
Creating table 'sbtest4'...
Inserting 150000 records into 'sbtest4'
Inserting 150000 records into 'sbtest3'
Creating a secondary index on 'sbtest3'...
Creating a secondary index on 'sbtest4'...
Creating table 'sbtest5'...
Creating table 'sbtest6'...
Inserting 150000 records into 'sbtest6'
Inserting 150000 records into 'sbtest5'
Creating a secondary index on 'sbtest6'...
Creating a secondary index on 'sbtest5'...
Creating table 'sbtest7'...
Creating table 'sbtest8'...
Inserting 150000 records into 'sbtest8'
Inserting 150000 records into 'sbtest7'
Creating a secondary index on 'sbtest8'...
Creating a secondary index on 'sbtest7'...

+-------------+----------------+----------------+-----------+------------+---------+------------+-------------------+
| Table Name  | Number of Rows | Storage Engine | Data Size | Index Size | Total   | ROW_FORMAT | TABLE_COLLATION   |
+-------------+----------------+----------------+-----------+------------+---------+------------+-------------------+
| sbt.sbtest1 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest2 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest3 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest4 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest5 | 148032 Rows    | InnoDB         | 32.56MB   | 0.00MB     | 32.56MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest6 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest7 | 148032 Rows    | InnoDB         | 32.56MB   | 0.00MB     | 32.56MB | Dynamic    | latin1_swedish_ci |
| sbt.sbtest8 | 150000 Rows    | InnoDB         | 0.02MB    | 0.00MB     | 0.02MB  | Dynamic    | latin1_swedish_ci |
+-------------+----------------+----------------+-----------+------------+---------+------------+-------------------+

sysbench mysql OLTP new benchmark:
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)

Running the test with following options:
Number of threads: 2
Report intermediate results every 1 second(s)
Initializing random number generator from seed (1).


Initializing worker threads...

Threads started!

[ 1s ] thds: 2 tps: 37.78 qps: 793.40 (r/w/o: 556.77/159.08/77.55) lat (ms,95%): 179.94 err/s: 0.00 reconn/s: 0.00
[ 2s ] thds: 2 tps: 203.48 qps: 4063.53 (r/w/o: 2848.68/807.89/406.95) lat (ms,95%): 26.68 err/s: 0.00 reconn/s: 0.00
[ 3s ] thds: 2 tps: 291.63 qps: 5823.60 (r/w/o: 4075.82/1164.52/583.26) lat (ms,95%): 13.22 err/s: 0.00 reconn/s: 0.00
[ 4s ] thds: 2 tps: 315.81 qps: 6327.19 (r/w/o: 4428.33/1267.24/631.62) lat (ms,95%): 12.52 err/s: 0.00 reconn/s: 0.00
[ 5s ] thds: 2 tps: 178.09 qps: 3537.75 (r/w/o: 2473.22/708.35/356.18) lat (ms,95%): 28.16 err/s: 0.00 reconn/s: 0.00
[ 6s ] thds: 2 tps: 203.57 qps: 4093.44 (r/w/o: 2869.00/817.29/407.15) lat (ms,95%): 21.89 err/s: 0.00 reconn/s: 0.00
[ 7s ] thds: 2 tps: 229.37 qps: 4571.39 (r/w/o: 3198.17/914.48/458.74) lat (ms,95%): 22.28 err/s: 0.00 reconn/s: 0.00
[ 8s ] thds: 2 tps: 164.44 qps: 3296.70 (r/w/o: 2310.08/657.75/328.87) lat (ms,95%): 25.28 err/s: 0.00 reconn/s: 0.00
[ 9s ] thds: 2 tps: 181.20 qps: 3618.97 (r/w/o: 2531.78/724.80/362.40) lat (ms,95%): 24.83 err/s: 0.00 reconn/s: 0.00
[ 10s ] thds: 2 tps: 112.97 qps: 2267.37 (r/w/o: 1589.56/451.87/225.94) lat (ms,95%): 40.37 err/s: 0.00 reconn/s: 0.00
[ 11s ] thds: 2 tps: 0.00 qps: 0.00 (r/w/o: 0.00/0.00/0.00) lat (ms,95%): 0.00 err/s: 0.00 reconn/s: 0.00
[ 12s ] thds: 2 tps: 19.98 qps: 392.64 (r/w/o: 272.75/79.93/39.96) lat (ms,95%): 1678.14 err/s: 0.00 reconn/s: 0.00
[ 13s ] thds: 2 tps: 61.05 qps: 1222.99 (r/w/o: 856.69/244.20/122.10) lat (ms,95%): 41.10 err/s: 0.00 reconn/s: 0.00
[ 14s ] thds: 2 tps: 100.96 qps: 2006.19 (r/w/o: 1400.43/403.84/201.92) lat (ms,95%): 30.26 err/s: 0.00 reconn/s: 0.00
[ 15s ] thds: 2 tps: 177.44 qps: 3568.82 (r/w/o: 2501.16/712.77/354.89) lat (ms,95%): 26.68 err/s: 0.00 reconn/s: 0.00
[ 16s ] thds: 2 tps: 160.57 qps: 3196.34 (r/w/o: 2235.93/639.27/321.14) lat (ms,95%): 33.12 err/s: 0.00 reconn/s: 0.00
[ 17s ] thds: 2 tps: 199.01 qps: 3990.27 (r/w/o: 2796.19/796.05/398.03) lat (ms,95%): 22.28 err/s: 0.00 reconn/s: 0.00
[ 18s ] thds: 2 tps: 189.95 qps: 3792.02 (r/w/o: 2651.32/760.80/379.90) lat (ms,95%): 31.94 err/s: 0.00 reconn/s: 0.00
[ 19s ] thds: 2 tps: 202.98 qps: 4075.58 (r/w/o: 2855.71/813.92/405.96) lat (ms,95%): 30.81 err/s: 0.00 reconn/s: 0.00
[ 20s ] thds: 2 tps: 129.01 qps: 2567.29 (r/w/o: 1796.21/513.06/258.03) lat (ms,95%): 32.53 err/s: 0.00 reconn/s: 0.00
[ 21s ] thds: 2 tps: 91.97 qps: 1846.43 (r/w/o: 1294.60/367.89/183.94) lat (ms,95%): 50.11 err/s: 0.00 reconn/s: 0.00
[ 22s ] thds: 2 tps: 128.85 qps: 2588.01 (r/w/o: 1806.91/523.40/257.70) lat (ms,95%): 35.59 err/s: 0.00 reconn/s: 0.00
[ 23s ] thds: 2 tps: 143.89 qps: 2871.89 (r/w/o: 2014.52/569.58/287.79) lat (ms,95%): 34.33 err/s: 0.00 reconn/s: 0.00
[ 24s ] thds: 2 tps: 152.33 qps: 3032.48 (r/w/o: 2120.53/607.30/304.65) lat (ms,95%): 29.72 err/s: 0.00 reconn/s: 0.00
[ 25s ] thds: 2 tps: 187.06 qps: 3758.27 (r/w/o: 2630.89/753.25/374.13) lat (ms,95%): 25.28 err/s: 0.00 reconn/s: 0.00
[ 26s ] thds: 2 tps: 203.35 qps: 4069.94 (r/w/o: 2846.86/816.38/406.69) lat (ms,95%): 23.52 err/s: 0.00 reconn/s: 0.00
[ 27s ] thds: 2 tps: 201.49 qps: 4011.80 (r/w/o: 2810.86/797.95/402.98) lat (ms,95%): 25.74 err/s: 0.00 reconn/s: 0.00
[ 28s ] thds: 2 tps: 212.17 qps: 4249.31 (r/w/o: 2976.32/848.66/424.33) lat (ms,95%): 19.29 err/s: 0.00 reconn/s: 0.00
[ 29s ] thds: 2 tps: 184.78 qps: 3699.61 (r/w/o: 2588.93/741.12/369.56) lat (ms,95%): 24.38 err/s: 0.00 reconn/s: 0.00
[ 30s ] thds: 2 tps: 153.79 qps: 3083.82 (r/w/o: 2155.08/621.16/307.58) lat (ms,95%): 31.94 err/s: 0.00 reconn/s: 0.00
SQL statistics:
    queries performed:
        read:                            67494
        write:                           19284
        other:                           9642
        total:                           96420
    transactions:                        4821   (160.62 per sec.)
    queries:                             96420  (3212.41 per sec.)
    ignored errors:                      0      (0.00 per sec.)
    reconnects:                          0      (0.00 per sec.)

General statistics:
    total time:                          30.0130s
    total number of events:              4821

Latency (ms):
         min:                                    1.54
         avg:                                   12.44
         max:                                 1952.60
         95th percentile:                       29.19
         sum:                                59990.70

Threads fairness:
    events (avg/stddev):           2410.5000/3.50
    execution time (avg/stddev):   29.9953/0.00


sysbench mysql OLTP new summary:
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
read: 67494
write: 19284
other: 9642
total: 96420
transactions/s: 160.62
queries/s: 3212.41
time: 30.0130s
min: 1.54
avg: 12.44
max: 1952.60
95th: 29.19

| mysql sysbench | sysbench | threads: | read: | write: | other: | total: | transactions/s: | queries/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| oltp_read_write.lua | 1.0.16 | 2 | 67494 | 19284 | 9642 | 96420 | 160.62 | 3212.41 | 30.0130s | 1.54 | 12.44 | 1952.60 | 29.19 |

sysbench,sysbench,threads,read,write,other,total,transactions/s,queries/s,time,min,avg,max,95th
oltp_read_write.lua,1.0.16,2,67494,19284,9642,96420,160.62,3212.41,30.0130s,1.54,12.44,1952.60,29.19

sysbench mysql cleanup database: sbt
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql cleanup
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)

Dropping table 'sbtest1'...
Dropping table 'sbtest2'...
Dropping table 'sbtest3'...
Dropping table 'sbtest4'...
Dropping table 'sbtest5'...
Dropping table 'sbtest6'...
Dropping table 'sbtest7'...
Dropping table 'sbtest8'...

2 CPU, 60GB US$15/month on older Intel Xeon E5-2630L v2 at 2.40GHz
cpu

Code (Text):

./sysbench.sh cpu

sysbench cpu --cpu-max-prime=20000 --threads=1 run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
prime: 20000
events/s: 261.47
time: 10.0029s
min: 3.57
avg: 3.82
max: 5.84
95th: 4.41

| cpu sysbench | threads: | events/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 1.0.16 | 1 | 261.47 | 10.0029s | 3.57 | 3.82 | 5.84 | 4.41 |

sysbench,threads,events/s,time,min,avg,max,95th
1.0.16,1,261.47,10.0029s,3.57,3.82,5.84,4.41

sysbench cpu --cpu-max-prime=20000 --threads=2 run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
prime: 20000
events/s: 515.88
time: 10.0020s
min: 3.57
avg: 3.87
max: 15.18
95th: 4.33

| cpu sysbench | threads: | events/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 1.0.16 | 2 | 515.88 | 10.0020s | 3.57 | 3.87 | 15.18 | 4.33 |

sysbench,threads,events/s,time,min,avg,max,95th
1.0.16,2,515.88,10.0020s,3.57,3.87,15.18,4.33

2 CPU, 60GB US$15/month on older Intel Xeon E5-2630L v2 at 2.40GHz
memory

Code (Text):

./sysbench.sh mem

sysbench memory --threads=1 --memory-block-size=1K --memory-scope=global --memory-total-size=1G --memory-oper=read run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
block-size: 1KiB
total-size: 1024MiB
operation: read
scope: global
total-ops: 1048576 (4107527.74 per second)
transferred (4011.26 MiB/sec)
time: 0.2530s
min: 0.00
avg: 0.00
max: 0.08
95th: 0.00

| memory sysbench | sysbench | threads: | block-size: | total-size: | operation: | total-ops: | transferred | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| memory | 1.0.16 | 1 | 1KiB | 1024MiB | read | 1048576 | 4011.26 | 0.2530s | 0.00 | 0.00 | 0.08 | 0.00 |

sysbench,sysbench,threads,block-size,total-size,operation,total-ops,transferred,time,min,avg,max,95th
memory,1.0.16,1,1KiB,1024MiB,read,1048576,4011.26,0.2530s,0.00,0.00,0.08,0.00

sysbench memory --threads=2 --memory-block-size=1K --memory-scope=global --memory-total-size=1G --memory-oper=read run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
block-size: 1KiB
total-size: 1024MiB
operation: read
scope: global
total-ops: 1048576 (6947617.96 per second)
transferred (6784.78 MiB/sec)
time: 0.1485s
min: 0.00
avg: 0.00
max: 0.19
95th: 0.00

| memory sysbench | sysbench | threads: | block-size: | total-size: | operation: | total-ops: | transferred | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| memory | 1.0.16 | 2 | 1KiB | 1024MiB | read | 1048576 | 6784.78 | 0.1485s | 0.00 | 0.00 | 0.19 | 0.00 |

sysbench,sysbench,threads,block-size,total-size,operation,total-ops,transferred,time,min,avg,max,95th
memory,1.0.16,2,1KiB,1024MiB,read,1048576,6784.78,0.1485s,0.00,0.00,0.19,0.00

2 CPU, 60GB US$15/month on older Intel Xeon E5-2630L v2 at 2.40GHz
file

Code (Text):

./sysbench.sh file

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqrd prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqrd-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Using synchronous I/O mode
Doing sequential read test
reads/s: 8166.90
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 31.90
written-MiB/s: 0.00
time: 10.0002s
min: 0.08
avg: 0.12
max: 18.22
95th: 0.16

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | read | 8166.90 | 0.00 | 0.00 | 31.90 | 0.00 | 10.0002s | 0.08 | 0.12 | 18.22 | 0.16 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,read,8166.90,0.00,0.00,31.90,0.00,10.0002s,0.08,0.12,18.22,0.16


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqwr prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqwr-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Using synchronous I/O mode
Doing sequential write (creation) test
reads/s: 0.00
writes/s: 2763.64
fsyncs/s: 3544.37
read-MiB/s: 0.00
written-MiB/s: 10.80
time: 10.0009s
min: 0.00
avg: 0.16
max: 18.59
95th: 0.41

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 2763.64 | 3544.37 | 0.00 | 10.80 | 10.0009s | 0.00 | 0.16 | 18.59 | 0.41 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,write,0.00,2763.64,3544.37,0.00,10.80,10.0009s,0.00,0.16,18.59,0.41


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndrd prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndrd-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random read test
reads/s: 3922.31
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 15.32
written-MiB/s: 0.00
time: 10.0002s
min: 0.08
avg: 0.25
max: 18.43
95th: 0.38

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | read | 3922.31 | 0.00 | 0.00 | 15.32 | 0.00 | 10.0002s | 0.08 | 0.25 | 18.43 | 0.38 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,read,3922.31,0.00,0.00,15.32,0.00,10.0002s,0.08,0.25,18.43,0.38


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndwr prepare

sysbench fileio --threads=1 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndwr-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random write test
reads/s: 0.00
writes/s: 5002.87
fsyncs/s: 6410.85
read-MiB/s: 0.00
written-MiB/s: 19.54
time: 10.0009s
min: 0.00
avg: 0.09
max: 41.92
95th: 0.20

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 5002.87 | 6410.85 | 0.00 | 19.54 | 10.0009s | 0.00 | 0.09 | 41.92 | 0.20 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,1,4KiB,I/O,write,0.00,5002.87,6410.85,0.00,19.54,10.0009s,0.00,0.09,41.92,0.20



sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqrd prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqrd-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Using synchronous I/O mode
Doing sequential read test
reads/s: 14746.43
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 57.60
written-MiB/s: 0.00
time: 10.0002s
min: 0.06
avg: 0.13
max: 23.20
95th: 0.18

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | read | 14746.43 | 0.00 | 0.00 | 57.60 | 0.00 | 10.0002s | 0.06 | 0.13 | 23.20 | 0.18 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,read,14746.43,0.00,0.00,57.60,0.00,10.0002s,0.06,0.13,23.20,0.18


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=seqwr prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=seqwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-seqwr-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Using synchronous I/O mode
Doing sequential write (creation) test
reads/s: 0.00
writes/s: 3167.97
fsyncs/s: 4068.82
read-MiB/s: 0.00
written-MiB/s: 12.37
time: 10.0015s
min: 0.00
avg: 0.28
max: 51.22
95th: 0.70

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 3167.97 | 4068.82 | 0.00 | 12.37 | 10.0015s | 0.00 | 0.28 | 51.22 | 0.70 |

sysbench,sysbench,threads,Block-size,synchronous,sequential,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,write,0.00,3167.97,4068.82,0.00,12.37,10.0015s,0.00,0.28,51.22,0.70


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndrd prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndrd --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndrd-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random read test
reads/s: 7993.39
writes/s: 0.00
fsyncs/s: 0.00
read-MiB/s: 31.22
written-MiB/s: 0.00
time: 10.0002s
min: 0.08
avg: 0.25
max: 19.42
95th: 0.34

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | read | 7993.39 | 0.00 | 0.00 | 31.22 | 0.00 | 10.0002s | 0.08 | 0.25 | 19.42 | 0.34 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,read,7993.39,0.00,0.00,31.22,0.00,10.0002s,0.08,0.25,19.42,0.34


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

sysbench fileio prepare
sysbench fileio --file-total-size=2048M --file-test-mode=rndwr prepare

sysbench fileio --threads=2 --file-num=128 --file-total-size=2048M --file-block-size=4096 --file-io-mode=sync --file-extra-flags=direct --file-test-mode=rndwr --time=10 --events=0 run
raw log saved: /home/sysbench/sysbench-fileio-rndwr-threads-2-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random write test
reads/s: 0.00
writes/s: 7759.27
fsyncs/s: 9955.67
read-MiB/s: 0.00
written-MiB/s: 30.31
time: 10.0006s
min: 0.00
avg: 0.11
max: 43.62
95th: 0.27

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 7759.27 | 9955.67 | 0.00 | 30.31 | 10.0006s | 0.00 | 0.11 | 43.62 | 0.27 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,95th
fileio,1.0.16,2,4KiB,I/O,write,0.00,7759.27,9955.67,0.00,30.31,10.0006s,0.00,0.11,43.62,0.27


sysbench fileio cleanup
sysbench fileio --file-total-size=2048M cleanup

| fileio sysbench | sysbench | threads: | Block-size | synchronous | sequential | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|fileio | 1.0.16 | 1 | 4KiB | I/O | read | 8166.90 | 0.00 | 0.00 | 31.90 | 0.00 | 10.0002s | 0.08 | 0.12 | 18.22 | 0.16 |
|fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 2763.64 | 3544.37 | 0.00 | 10.80 | 10.0009s | 0.00 | 0.16 | 18.59 | 0.41 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | read | 14746.43 | 0.00 | 0.00 | 57.60 | 0.00 | 10.0002s | 0.06 | 0.13 | 23.20 | 0.18 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 3167.97 | 4068.82 | 0.00 | 12.37 | 10.0015s | 0.00 | 0.28 | 51.22 | 0.70 |

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|fileio | 1.0.16 | 1 | 4KiB | I/O | read | 3922.31 | 0.00 | 0.00 | 15.32 | 0.00 | 10.0002s | 0.08 | 0.25 | 18.43 | 0.38 |
|fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 5002.87 | 6410.85 | 0.00 | 19.54 | 10.0009s | 0.00 | 0.09 | 41.92 | 0.20 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | read | 7993.39 | 0.00 | 0.00 | 31.22 | 0.00 | 10.0002s | 0.08 | 0.25 | 19.42 | 0.34 |
|fileio | 1.0.16 | 2 | 4KiB | I/O | write | 0.00 | 7759.27 | 9955.67 | 0.00 | 30.31 | 10.0006s | 0.00 | 0.11 | 43.62 | 0.27 |

2 CPU, 60GB US$15/month on older Intel Xeon E5-2630L v2 at 2.40GHz
file-fsync

Code (Text):

./sysbench.sh file-fsync

sysbench fileio fsync prepare
sysbench fileio --time=30 --file-num=1 --file-extra-flags= --file-total-size=4096 --file-block-size=4096 --file-fsync-all=on --file-test-mode=rndwr --file-fsync-freq=0 --file-fsync-end=0 --threads=1 --percentile=99 prepare

sysbench fileio --threads=1 --time=30 --file-num=1 --file-extra-flags= --file-total-size=4096 --file-block-size=4096 --file-fsync-all=on --file-test-mode=rndwr --file-fsync-freq=0 --file-fsync-end=0 --percentile=99 run
raw log saved: /home/sysbench/sysbench-fileio-fsync-threads-1-raw.log

sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 1
Block-size 4KiB
Read/Write ratio for combined random IO test: 1.50
Using synchronous I/O mode
Doing random write test
reads/s: 0.00
writes/s: 3663.67
fsyncs/s: 3663.67
read-MiB/s: 0.00
written-MiB/s: 14.31
time: 30.0002s
min: 0.11
avg: 0.27
max: 18.76
99th: 0.77

| fileio sysbench | sysbench | threads: | Block-size | synchronous | random | reads/s: | writes/s: | fsyncs/s: | read-MiB/s: | written-MiB/s: | time: | min: | avg: | max: | 99th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| fileio | 1.0.16 | 1 | 4KiB | I/O | write | 0.00 | 3663.67 | 3663.67 | 0.00 | 14.31 | 30.0002s | 0.11 | 0.27 | 18.76 | 0.77 |

sysbench,sysbench,threads,Block-size,synchronous,random,reads/s,writes/s,fsyncs/s,read-MiB/s,written-MiB/s,time,min,avg,max,99th
fileio,1.0.16,1,4KiB,I/O,write,0.00,3663.67,3663.67,0.00,14.31,30.0002s,0.11,0.27,18.76,0.77


sysbench fileio cleanup
sysbench fileio --file-num=1 --file-extra-flags= --file-total-size=4096 --file-block-size=4096 cleanup

2 CPU, 60GB US$15/month on older Intel Xeon E5-2630L v2 at 2.40GHz
mysql oltp

Code (Text):

./sysbench.sh mysqloltpnew

MySQL Buffers
aria_pagecache_buffer_size           16777216
aria_sort_buffer_size                16777216
bulk_insert_buffer_size              8388608
innodb_buffer_pool_dump_at_shutdown  OFF
innodb_buffer_pool_dump_now          OFF
innodb_buffer_pool_dump_pct          100
innodb_buffer_pool_filename          ib_buffer_pool
innodb_buffer_pool_instances         8
innodb_buffer_pool_load_abort        OFF
innodb_buffer_pool_load_at_startup   OFF
innodb_buffer_pool_load_now          OFF
innodb_buffer_pool_populate          OFF
innodb_buffer_pool_size              268435456
innodb_change_buffer_max_size        25
innodb_change_buffering              all
innodb_log_buffer_size               8388608
innodb_sort_buffer_size              2097152
join_buffer_size                     262144
join_buffer_space_limit              2097152
key_buffer_size                      33554432
mrr_buffer_size                      262144
myisam_sort_buffer_size              134217728
net_buffer_length                    16384
optimizer_switch                     index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on
preload_buffer_size                  32768
read_buffer_size                     262144
read_rnd_buffer_size                 262144
sort_buffer_size                     262144
sql_buffer_result                    OFF

MySQL Limits
aria_pagecache_division_limit         100
delayed_insert_limit                  100
expensive_subquery_limit              100
ft_query_expansion_limit              20
innodb_ft_result_cache_limit          2000000000
join_buffer_space_limit               2097152
key_cache_division_limit              100
log_slow_rate_limit                   1
min_examined_row_limit                0
open_files_limit                      8554
optimizer_selectivity_sampling_limit  100
query_cache_limit                     0
relay_log_space_limit                 0
sql_select_limit                      18446744073709551615
thread_pool_stall_limit               500
updatable_views_with_limit            YES

MySQL Maxes
aria_max_sort_file_size           9223372036853727232
extra_max_connections             2
ft_max_word_len                   84
group_concat_max_len              1024
innodb_adaptive_max_sleep_delay   150000
innodb_compression_pad_pct_max    50
innodb_file_format_max            Antelope
innodb_ft_max_token_size          84
innodb_io_capacity_max            4000
innodb_max_bitmap_file_size       104857600
innodb_max_changed_pages          1000000
innodb_max_dirty_pages_pct        75.000000
innodb_max_dirty_pages_pct_lwm    0.001000
innodb_max_purge_lag              0
innodb_max_purge_lag_delay        0
innodb_online_alter_log_max_size  134217728
max_allowed_packet                33554432
max_binlog_cache_size             18446744073709547520
max_binlog_size                   1073741824
max_binlog_stmt_cache_size        18446744073709547520
max_connect_errors                100000
max_connections                   350
max_delayed_threads               20
max_digest_length                 1024
max_error_count                   64
max_heap_table_size               67108864
max_insert_delayed_threads        20
max_join_size                     18446744073709551615
max_length_for_sort_data          1024
max_long_data_size                33554432
max_prepared_stmt_count           16382
max_relay_log_size                1073741824
max_seeks_for_key                 1000
max_sort_length                   1024
max_sp_recursion_depth            0
max_tmp_tables                    32
max_user_connections              0
max_write_lock_count              4294967295
myisam_max_sort_file_size         2147483648
slave_max_allowed_packet          1073741824
slave_parallel_max_queued         131072
thread_pool_max_threads           500

MySQL Concurrency
concurrent_insert           ALWAYS
innodb_commit_concurrency   0
innodb_concurrency_tickets  5000
innodb_thread_concurrency   0
thread_concurrency          10

MySQL Read/Write
delay_key_write           ON
innodb_doublewrite        ON
innodb_use_atomic_writes  OFF
max_write_lock_count      4294967295
net_write_timeout         60

MySQL Threads
aria_repair_threads                      1
binlog_optimize_thread_scheduling        ON
debug_no_thread_alarm                    OFF
innodb_purge_threads                     1
innodb_read_io_threads                   4
innodb_thread_concurrency                0
innodb_thread_sleep_delay                10000
innodb_write_io_threads                  2
max_delayed_threads                      20
max_insert_delayed_threads               20
myisam_repair_threads                    1
performance_schema_max_thread_classes    50
performance_schema_max_thread_instances  -1
slave_domain_parallel_threads            0
slave_parallel_threads                   0
thread_cache_size                        64
thread_concurrency                       10
thread_handling                          pool-of-threads
thread_pool_idle_timeout                 60
thread_pool_max_threads                  500
thread_pool_oversubscribe                3
thread_pool_size                         500
thread_pool_stall_limit                  500
thread_stack                             294912

MySQL Binlog
binlog_annotate_row_events               OFF
binlog_cache_size                        32768
binlog_checksum                          NONE
binlog_commit_wait_count                 0
binlog_commit_wait_usec                  100000
binlog_direct_non_transactional_updates  OFF
binlog_format                            STATEMENT
binlog_optimize_thread_scheduling        ON
binlog_stmt_cache_size                   32768
gtid_binlog_pos
gtid_binlog_state
innodb_api_enable_binlog                 OFF
innodb_locks_unsafe_for_binlog           OFF
log_bin                                  OFF
log_bin_trust_function_creators          OFF
max_binlog_cache_size                    18446744073709547520
max_binlog_size                          1073741824
max_binlog_stmt_cache_size               18446744073709547520
sql_log_bin                              ON
sync_binlog                              0

MySQL InnoDB
ignore_builtin_innodb                      OFF
innodb_adaptive_flushing                   ON
innodb_adaptive_flushing_lwm               10.000000
innodb_adaptive_hash_index                 ON
innodb_adaptive_hash_index_partitions      1
innodb_adaptive_max_sleep_delay            150000
innodb_additional_mem_pool_size            8388608
innodb_api_bk_commit_interval              5
innodb_api_disable_rowlock                 OFF
innodb_api_enable_binlog                   OFF
innodb_api_enable_mdl                      OFF
innodb_api_trx_level                       0
innodb_autoextend_increment                64
innodb_autoinc_lock_mode                   1
innodb_buffer_pool_dump_at_shutdown        OFF
innodb_buffer_pool_dump_now                OFF
innodb_buffer_pool_dump_pct                100
innodb_buffer_pool_filename                ib_buffer_pool
innodb_buffer_pool_instances               8
innodb_buffer_pool_load_abort              OFF
innodb_buffer_pool_load_at_startup         OFF
innodb_buffer_pool_load_now                OFF
innodb_buffer_pool_populate                OFF
innodb_buffer_pool_size                    268435456
innodb_change_buffer_max_size              25
innodb_change_buffering                    all
innodb_checksum_algorithm                  innodb
innodb_checksums                           ON
innodb_cleaner_lsn_age_factor              high_checkpoint
innodb_cmp_per_index_enabled               OFF
innodb_commit_concurrency                  0
innodb_compression_failure_threshold_pct   5
innodb_compression_level                   6
innodb_compression_pad_pct_max             50
innodb_concurrency_tickets                 5000
innodb_corrupt_table_action                assert
innodb_data_file_path                      ibdata1:10M:autoextend
innodb_data_home_dir
innodb_disable_sort_file_cache             OFF
innodb_doublewrite                         ON
innodb_empty_free_list_algorithm           backoff
innodb_fake_changes                        OFF
innodb_fast_shutdown                       1
innodb_file_format                         Barracuda
innodb_file_format_check                   ON
innodb_file_format_max                     Antelope
innodb_file_per_table                      ON
innodb_flush_log_at_timeout                1
innodb_flush_log_at_trx_commit             2
innodb_flush_method                        O_DIRECT
innodb_flush_neighbors                     1
innodb_flushing_avg_loops                  30
innodb_force_load_corrupted                OFF
innodb_force_recovery                      0
innodb_foreground_preflush                 exponential_backoff
innodb_ft_aux_table
innodb_ft_cache_size                       8000000
innodb_ft_enable_diag_print                OFF
innodb_ft_enable_stopword                  ON
innodb_ft_max_token_size                   84
innodb_ft_min_token_size                   3
innodb_ft_num_word_optimize                2000
innodb_ft_result_cache_limit               2000000000
innodb_ft_server_stopword_table
innodb_ft_sort_pll_degree                  2
innodb_ft_total_cache_size                 640000000
innodb_ft_user_stopword_table
innodb_io_capacity                         2000
innodb_io_capacity_max                     4000
innodb_kill_idle_transaction               0
innodb_large_prefix                        OFF
innodb_lock_wait_timeout                   50
innodb_locking_fake_changes                ON
innodb_locks_unsafe_for_binlog             OFF
innodb_log_arch_dir                        ./
innodb_log_arch_expire_sec                 0
innodb_log_archive                         OFF
innodb_log_block_size                      512
innodb_log_buffer_size                     8388608
innodb_log_checksum_algorithm              innodb
innodb_log_compressed_pages                ON
innodb_log_file_size                       134217728
innodb_log_files_in_group                  2
innodb_log_group_home_dir                  ./
innodb_lru_scan_depth                      1024
innodb_max_bitmap_file_size                104857600
innodb_max_changed_pages                   1000000
innodb_max_dirty_pages_pct                 75.000000
innodb_max_dirty_pages_pct_lwm             0.001000
innodb_max_purge_lag                       0
innodb_max_purge_lag_delay                 0
innodb_mirrored_log_groups                 1
innodb_monitor_disable
innodb_monitor_enable
innodb_monitor_reset
innodb_monitor_reset_all
innodb_old_blocks_pct                      37
innodb_old_blocks_time                     1000
innodb_online_alter_log_max_size           134217728
innodb_open_files                          1000
innodb_optimize_fulltext_only              OFF
innodb_page_size                           16384
innodb_print_all_deadlocks                 OFF
innodb_print_lock_wait_timeout_info        OFF
innodb_purge_batch_size                    300
innodb_purge_threads                       1
innodb_random_read_ahead                   OFF
innodb_read_ahead_threshold                56
innodb_read_io_threads                     4
innodb_read_only                           OFF
innodb_replication_delay                   0
innodb_rollback_on_timeout                 OFF
innodb_rollback_segments                   128
innodb_sched_priority_cleaner              19
innodb_show_locks_held                     10
innodb_show_verbose_locks                  0
innodb_simulate_comp_failures              0
innodb_sort_buffer_size                    2097152
innodb_spin_wait_delay                     6
innodb_stats_auto_recalc                   ON
innodb_stats_include_delete_marked         OFF
innodb_stats_method                        nulls_equal
innodb_stats_modified_counter              0
innodb_stats_on_metadata                   OFF
innodb_stats_persistent                    ON
innodb_stats_persistent_sample_pages       20
innodb_stats_sample_pages                  8
innodb_stats_traditional                   ON
innodb_stats_transient_sample_pages        8
innodb_status_output                       OFF
innodb_status_output_locks                 OFF
innodb_strict_mode                         OFF
innodb_support_xa                          ON
innodb_sync_array_size                     1
innodb_sync_spin_loops                     30
innodb_table_locks                         ON
innodb_thread_concurrency                  0
innodb_thread_sleep_delay                  10000
innodb_tmpdir
innodb_track_changed_pages                 OFF
innodb_undo_directory                      .
innodb_undo_logs                           128
innodb_undo_tablespaces                    0
innodb_use_atomic_writes                   OFF
innodb_use_fallocate                       OFF
innodb_use_global_flush_log_at_trx_commit  ON
innodb_use_native_aio                      ON
innodb_use_stacktrace                      OFF
innodb_use_sys_malloc                      ON
innodb_version                             5.6.39-84.1
innodb_write_io_threads                    2


sysbench prepare database: sbt
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql prepare
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)

Initializing worker threads...

Creating table 'sbtest2'...
Creating table 'sbtest1'...
Inserting 150000 records into 'sbtest2'
Inserting 150000 records into 'sbtest1'
Creating a secondary index on 'sbtest2'...
Creating a secondary index on 'sbtest1'...
Creating table 'sbtest3'...
Inserting 150000 records into 'sbtest3'
Creating table 'sbtest4'...
Inserting 150000 records into 'sbtest4'
Creating a secondary index on 'sbtest3'...
Creating a secondary index on 'sbtest4'...
Creating table 'sbtest5'...
Inserting 150000 records into 'sbtest5'
Creating table 'sbtest6'...
Inserting 150000 records into 'sbtest6'
Creating a secondary index on 'sbtest5'...
Creating a secondary index on 'sbtest6'...
Creating table 'sbtest7'...
Inserting 150000 records into 'sbtest7'
Creating table 'sbtest8'...
Inserting 150000 records into 'sbtest8'
Creating a secondary index on 'sbtest7'...
Creating a secondary index on 'sbtest8'...

+-------------+----------------+----------------+-----------+------------+---------+------------+-------------------+
| Table Name  | Number of Rows | Storage Engine | Data Size | Index Size | Total   | ROW_FORMAT | TABLE_COLLATION   |
+-------------+----------------+----------------+-----------+------------+---------+------------+-------------------+
| sbt.sbtest1 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Compact    | latin1_swedish_ci |
| sbt.sbtest2 | 150000 Rows    | InnoDB         | 0.30MB    | 0.00MB     | 0.30MB  | Compact    | latin1_swedish_ci |
| sbt.sbtest3 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Compact    | latin1_swedish_ci |
| sbt.sbtest4 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Compact    | latin1_swedish_ci |
| sbt.sbtest5 | 148032 Rows    | InnoDB         | 32.56MB   | 2.52MB     | 35.08MB | Compact    | latin1_swedish_ci |
| sbt.sbtest6 | 150000 Rows    | InnoDB         | 0.02MB    | 0.00MB     | 0.02MB  | Compact    | latin1_swedish_ci |
| sbt.sbtest7 | 150000 Rows    | InnoDB         | 0.02MB    | 0.00MB     | 0.02MB  | Compact    | latin1_swedish_ci |
| sbt.sbtest8 | 150000 Rows    | InnoDB         | 0.02MB    | 0.00MB     | 0.02MB  | Compact    | latin1_swedish_ci |
+-------------+----------------+----------------+-----------+------------+---------+------------+-------------------+

sysbench mysql OLTP new benchmark:
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)

Running the test with following options:
Number of threads: 2
Report intermediate results every 1 second(s)
Initializing random number generator from seed (1).


Initializing worker threads...

Threads started!

[ 1s ] thds: 2 tps: 256.23 qps: 5155.55 (r/w/o: 3612.18/1028.92/514.46) lat (ms,95%): 16.12 err/s: 0.00 reconn/s: 0.00
[ 2s ] thds: 2 tps: 277.19 qps: 5538.87 (r/w/o: 3877.71/1106.77/554.39) lat (ms,95%): 12.75 err/s: 0.00 reconn/s: 0.00
[ 3s ] thds: 2 tps: 318.02 qps: 6358.44 (r/w/o: 4452.31/1270.09/636.04) lat (ms,95%): 9.91 err/s: 0.00 reconn/s: 0.00
[ 4s ] thds: 2 tps: 317.03 qps: 6342.66 (r/w/o: 4436.46/1272.13/634.07) lat (ms,95%): 8.90 err/s: 0.00 reconn/s: 0.00
[ 5s ] thds: 2 tps: 309.99 qps: 6180.83 (r/w/o: 4324.88/1235.97/619.98) lat (ms,95%): 9.56 err/s: 0.00 reconn/s: 0.00
[ 6s ] thds: 2 tps: 296.02 qps: 5943.31 (r/w/o: 4163.22/1188.06/592.03) lat (ms,95%): 9.06 err/s: 0.00 reconn/s: 0.00
[ 7s ] thds: 2 tps: 281.99 qps: 5628.78 (r/w/o: 3940.84/1123.96/563.98) lat (ms,95%): 11.65 err/s: 0.00 reconn/s: 0.00
[ 8s ] thds: 2 tps: 291.99 qps: 5834.72 (r/w/o: 4082.80/1167.94/583.97) lat (ms,95%): 9.22 err/s: 0.00 reconn/s: 0.00
[ 9s ] thds: 2 tps: 292.01 qps: 5858.21 (r/w/o: 4103.15/1171.04/584.02) lat (ms,95%): 8.13 err/s: 0.00 reconn/s: 0.00
[ 10s ] thds: 2 tps: 286.99 qps: 5741.82 (r/w/o: 4018.88/1148.96/573.98) lat (ms,95%): 10.46 err/s: 0.00 reconn/s: 0.00
[ 11s ] thds: 2 tps: 281.01 qps: 5608.13 (r/w/o: 3923.09/1123.03/562.01) lat (ms,95%): 7.84 err/s: 0.00 reconn/s: 0.00
[ 12s ] thds: 2 tps: 254.02 qps: 5065.35 (r/w/o: 3544.24/1013.07/508.03) lat (ms,95%): 8.43 err/s: 0.00 reconn/s: 0.00
[ 13s ] thds: 2 tps: 236.38 qps: 4734.65 (r/w/o: 3316.35/945.53/472.77) lat (ms,95%): 12.30 err/s: 0.00 reconn/s: 0.00
[ 14s ] thds: 2 tps: 291.73 qps: 5847.62 (r/w/o: 4097.24/1166.92/583.46) lat (ms,95%): 11.45 err/s: 0.00 reconn/s: 0.00
[ 15s ] thds: 2 tps: 305.01 qps: 6098.24 (r/w/o: 4264.16/1224.05/610.02) lat (ms,95%): 11.65 err/s: 0.00 reconn/s: 0.00
[ 16s ] thds: 2 tps: 331.98 qps: 6637.59 (r/w/o: 4645.71/1327.92/663.96) lat (ms,95%): 9.56 err/s: 0.00 reconn/s: 0.00
[ 17s ] thds: 2 tps: 347.00 qps: 6937.05 (r/w/o: 4856.03/1387.01/694.00) lat (ms,95%): 8.90 err/s: 0.00 reconn/s: 0.00
[ 18s ] thds: 2 tps: 333.96 qps: 6684.30 (r/w/o: 4683.51/1332.86/667.93) lat (ms,95%): 9.73 err/s: 0.00 reconn/s: 0.00
[ 19s ] thds: 2 tps: 336.99 qps: 6742.80 (r/w/o: 4720.86/1347.96/673.98) lat (ms,95%): 10.84 err/s: 0.00 reconn/s: 0.00
[ 20s ] thds: 2 tps: 295.07 qps: 5903.38 (r/w/o: 4132.96/1180.28/590.14) lat (ms,95%): 11.24 err/s: 0.00 reconn/s: 0.00
[ 21s ] thds: 2 tps: 333.92 qps: 6657.43 (r/w/o: 4653.90/1335.69/667.84) lat (ms,95%): 11.45 err/s: 0.00 reconn/s: 0.00
[ 22s ] thds: 2 tps: 312.05 qps: 6269.95 (r/w/o: 4389.66/1256.19/624.09) lat (ms,95%): 11.87 err/s: 0.00 reconn/s: 0.00
[ 23s ] thds: 2 tps: 317.00 qps: 6318.93 (r/w/o: 4423.95/1260.99/633.99) lat (ms,95%): 10.09 err/s: 0.00 reconn/s: 0.00
[ 24s ] thds: 2 tps: 337.04 qps: 6738.87 (r/w/o: 4717.61/1347.17/674.09) lat (ms,95%): 10.09 err/s: 0.00 reconn/s: 0.00
[ 25s ] thds: 2 tps: 340.98 qps: 6824.52 (r/w/o: 4774.66/1367.90/681.95) lat (ms,95%): 9.22 err/s: 0.00 reconn/s: 0.00
[ 26s ] thds: 2 tps: 326.01 qps: 6511.13 (r/w/o: 4559.09/1300.03/652.01) lat (ms,95%): 10.46 err/s: 0.00 reconn/s: 0.00
[ 27s ] thds: 2 tps: 344.99 qps: 6918.72 (r/w/o: 4848.80/1379.94/689.97) lat (ms,95%): 9.06 err/s: 0.00 reconn/s: 0.00
[ 28s ] thds: 2 tps: 357.95 qps: 7145.00 (r/w/o: 4997.30/1431.80/715.90) lat (ms,95%): 9.06 err/s: 0.00 reconn/s: 0.00
[ 29s ] thds: 2 tps: 351.84 qps: 7047.85 (r/w/o: 4936.80/1407.37/703.69) lat (ms,95%): 8.58 err/s: 0.00 reconn/s: 0.00
[ 30s ] thds: 2 tps: 327.96 qps: 6532.11 (r/w/o: 4566.38/1311.82/653.91) lat (ms,95%): 13.95 err/s: 0.00 reconn/s: 0.00
SQL statistics:
    queries performed:
        read:                            130088
        write:                           37168
        other:                           18584
        total:                           185840
    transactions:                        9292   (309.54 per sec.)
    queries:                             185840 (6190.75 per sec.)
    ignored errors:                      0      (0.00 per sec.)
    reconnects:                          0      (0.00 per sec.)

General statistics:
    total time:                          30.0169s
    total number of events:              9292

Latency (ms):
         min:                                    2.69
         avg:                                    6.45
         max:                                  433.00
         95th percentile:                       10.46
         sum:                                59977.38

Threads fairness:
    events (avg/stddev):           4646.0000/37.00
    execution time (avg/stddev):   29.9887/0.01


sysbench mysql OLTP new summary:
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql run
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)
threads: 2
read: 130088
write: 37168
other: 18584
total: 185840
transactions/s: 309.54
queries/s: 6190.75
time: 30.0169s
min: 2.69
avg: 6.45
max: 433.00
95th: 10.46

| mysql sysbench | sysbench | threads: | read: | write: | other: | total: | transactions/s: | queries/s: | time: | min: | avg: | max: | 95th: |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| oltp_read_write.lua | 1.0.16 | 2 | 130088 | 37168 | 18584 | 185840 | 309.54 | 6190.75 | 30.0169s | 2.69 | 6.45 | 433.00 | 10.46 |

sysbench,sysbench,threads,read,write,other,total,transactions/s,queries/s,time,min,avg,max,95th
oltp_read_write.lua,1.0.16,2,130088,37168,18584,185840,309.54,6190.75,30.0169s,2.69,6.45,433.00,10.46

sysbench mysql cleanup database: sbt
sysbench oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-socket=/var/lib/mysql/mysql.sock --mysql-user=sbtest --mysql-password=sbtestpass --mysql-db=sbt --mysql-storage-engine=InnoDB --time=30 --threads=2 --report-interval=1 --rand-type=uniform --rand-seed=1 --table-size=150000 --tables=8 --db-driver=mysql cleanup
sysbench 1.0.16 (using bundled LuaJIT 2.1.0-beta2)

Dropping table 'sbtest1'...
Dropping table 'sbtest2'...
Dropping table 'sbtest3'...
Dropping table 'sbtest4'...
Dropping table 'sbtest5'...
Dropping table 'sbtest6'...
Dropping table 'sbtest7'...
Dropping table 'sbtest8'...

Conclusion

I was surprised by the sysbench benchmark results on the DigitalOcean droplets. I expected the newer faster cpu based droplet with Intel Xeon Gold 6140 Skylake cpus to be much faster all round. Instead, it was truly just faster cpu and memory performance paired with much lower disk I/O (fsync, random and sequential) and thus lower MySQL performance in terms of both throughput and latency response times.

It maybe just isolated to my droplets disk I/O as the test sample size is small. So you should always test your droplets after you deploy them to see if their performance is what you expect and suited for your particular task i.e cpu and/or memory intensive or disk I/O and MySQL related performance.

I am limited by further extensive benchmarks as my original SFO1 droplet is a live production site though. So sysbench.sh runs are a form of quicker testing to do.

Hope you enjoyed by benchmarks, you can sign up for DigitalOcean via my affiliate link and get yourself US$100 free credit on new signups to play with too over 60 days. Then once you spent US$25, I will get US$25 referral credit too

eva2000 · Dec 17, 2018

Nginx HTTP/2 HTTPS Benchmarks

I also ran my custom Nginx HTTP/2 HTTPS benchmark too, https_bench.sh and definitely newer Intel Xeon Gold 6140 was much faster than Intel Xeon E5-2630Lv2 especially for Nginx HTTP/2 HTTPS with Nginx Brotli compression speedup compared to with Gzip compression.

Once Centmin Mod 123.09beta01 installed, run https_bench.sh command

If you want to run on existing Centmin Mod Nginx installs an extended Nginx HTTP/2 HTTPS RSA 2048 bit + ECDSA 256bit SSL certificated based h2load tests, you can use the below commands.

h2load HTTP/2 HTTPS load test tool will be run single threaded so tests only 1 cpu core of the system.

https_bench.sh will create a dummy Nginx HTTP/2 HTTPS vhost called http2.domain.com and setup both self-signed RSA 2048bit and ECDSA 256bit SSL certificates on the dummy Nginx vhost to run h2load HTTP/2 HTTPS load tests against. After https_bench.sh completes, it will automatically remove the http2.domain.com dummy Nginx vhost site and all SSL certificates.

The Nginx HTTP/2 HTTPS h2load tests will test 4 sets of SSL Ciphers for ECDHE-RSA-AES128-GCM-SHA256, ECDHE-RSA-AES256-GCM-SHA384, ECDHE-ECDSA-AES128-GCM-SHA256 and ECDHE-ECDSA-AES256-GCM-SHA384.

This test will test gzip (and brotli if Nginx support is detected) HTTP compression load tests using h2load HTTP/2 HTTPS tester tool.

All of the scripts output and benchmark results will be logged into a file at /root/centminlogs/h2load-nginx-https-${DT}.log where ${DT} is date timestamp.

h2load HTTP/2 HTTPS load test configurations

There's 8x h2load test configurations in total. h2load HTTP/2 HTTPS load test tool will be run single threaded so tests only 1 cpu core of the system:

h2load --ciphers=ECDHE-RSA-AES128-GCM-SHA256 -H 'Accept-Encoding: gzip' -c100 -n1000 https://http2.domain.com

h2load --ciphers=ECDHE-RSA-AES256-GCM-SHA384 -H 'Accept-Encoding: gzip' -c100 -n1000 https://http2.domain.com

h2load --ciphers=ECDHE-ECDSA-AES128-GCM-SHA256 -H 'Accept-Encoding: gzip' -c100 -n1000 https://http2.domain.com

h2load --ciphers=ECDHE-ECDSA-AES256-GCM-SHA384 -H 'Accept-Encoding: gzip' -c100 -n1000 https://http2.domain.com

h2load --ciphers=ECDHE-RSA-AES128-GCM-SHA256 -H 'Accept-Encoding: gzip' -c300 -n6000 https://http2.domain.com

h2load --ciphers=ECDHE-RSA-AES256-GCM-SHA384 -H 'Accept-Encoding: gzip' -c300 -n6000 https://http2.domain.com

h2load --ciphers=ECDHE-ECDSA-AES128-GCM-SHA256 -H 'Accept-Encoding: gzip' -c300 -n6000 https://http2.domain.com

h2load --ciphers=ECDHE-ECDSA-AES256-GCM-SHA384 -H 'Accept-Encoding: gzip' -c300 -n6000 https://http2.domain.com

Install Commands:

Note: if you already have a Nginx vhost called http2.domain.com, you can edit /root/tools/https_bench.sh script's variable vhostname=http2.domain.com by overriding it in a separately created file persistent config file at /root/tools/https_bench.ini before running https_bench.sh.
Code (Text):
mkdir -p /root/tools
cd /root/tools
wget -O https_bench.sh https://github.com/centminmod/centminmodbench/raw/master/https_bench.sh
chmod +x https_bench.sh
time /root/tools/https_bench.sh
To update script
Code (Text):
wget -O /root/tools/https_bench.sh https://github.com/centminmod/centminmodbench/raw/master/https_bench.sh
Results focusing on 300 concurrent users and 6000 requests h2load HTTP/2 HTTPS tests. Intel Xeon Gold 6140 seems to have accelerated Nginx brotli compression based HTTP/2 HTTPS performance alot especially when paired with ECDSA 256bit SSL certificate based Nginx HTTP/2 HTTPS configurations - Intel Xeon Gold 6140 was up to 2.09x to 2.88x times faster than Intel Xeon E5-2630L v2 !

new 2 cpu droplet Intel Xeon Gold 6140 at 2.30Ghz,
- ECDHE-RSA-AES128-GCM-SHA256 cipher for RSA 2048bit SSL cert with Nginx HTTP/2 HTTPS gzip = 5,739.50 req/s
- ECDHE-ECDSA-AES128-GCM-SHA256 cipher for ECDSA 256bit SSL cert with Nginx HTTP/2 HTTPS gzip = 6,550.23 req/s - newer Intel generation cpus have much improved ECDSA ssl cipher performance so it's expected that ECDSA 256bit SSL cert based HTTP/2 HTTPS scales better than standard RSA 2048bit SSL certs.
- ECDHE-RSA-AES128-GCM-SHA256 cipher for RSA 2048bit SSL cert with Nginx HTTP/2 HTTPS brotli compression = 8,371.04 req/s
- ECDHE-ECDSA-AES128-GCM-SHA256 cipher for ECDSA 256bit SSL cert with Nginx HTTP/2 HTTPS brotli compression = 9,549.73 req/s - newer Intel generation cpus have much improved HTTP/2 HTTPS performance with Nginx brotli compression compared to the older cpu.

old 2 cpu droplet Intel Xeon E5-2630L v2 at 2.40GHz,
- ECDHE-RSA-AES128-GCM-SHA256 cipher for RSA 2048bit SSL cert with Nginx HTTP/2 HTTPS gzip = 3,509.22 req/s
- ECDHE-ECDSA-AES128-GCM-SHA256 cipher for ECDSA 256bit SSL cert with Nginx HTTP/2 HTTPS gzip = 3,866.01 req/s
- ECDHE-RSA-AES128-GCM-SHA256 cipher for RSA 2048bit SSL cert with Nginx HTTP/2 HTTPS brotli compression = 3,783.44 req/s
- ECDHE-ECDSA-AES128-GCM-SHA256 cipher for ECDSA 256bit SSL cert with Nginx HTTP/2 HTTPS brotli compression = 4,556.93 req/s - newer Intel generation cpus have much improved HTTP/2 HTTPS performance with Nginx brotli compression compared to the older cpu.

Full raw numbers

new 2 cpu droplet Intel Xeon Gold 6140 at 2.30Ghz
Code (Text):
users requests req/s encoding cipher protocol started succeeded
100 1000 3734.80 gzip ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 5348.74 gzip ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
100 1000 4782.31 gzip ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 5013.96 gzip ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
300 6000 5739.50 gzip ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 6241.76 gzip ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
300 6000 6550.23 gzip ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 5351.88 gzip ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
100 1000 5233.82 br'  ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 6770.57 br'  ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
100 1000 6107.69 br'  ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 7612.96 br'  ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
300 6000 8371.04 br'  ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 7802.17 br'  ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
300 6000 9549.73 br'  ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 11325.97 br'  ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
-------------------------------------------------------------------------------------------
h2load result summary
min:      avg:      max:       stddev:   requests-succeeded:
3734.800  6596.071  11325.970  1946.019  100.00
-------------------------------------------------------------------------------------------
old 2 cpu droplet Intel Xeon E5-2630L v2 at 2.40GHz
Code (Text):
users requests req/s encoding cipher protocol started succeeded
100 1000 2428.16 gzip ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 2394.00 gzip ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
100 1000 3131.49 gzip ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 3103.06 gzip ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
300 6000 3509.22 gzip ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 3522.34 gzip ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
300 6000 3866.01 gzip ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 3784.18 gzip ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
100 1000 2777.54 br'  ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 2935.89 br'  ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
100 1000 3441.50 br'  ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 1000 1000
100 1000 3511.08 br'  ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 1000 1000
300 6000 3783.44 br'  ECDHE-RSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 3590.06 br'  ECDHE-RSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
300 6000 4556.93 br'  ECDHE-ECDSA-AES128-GCM-SHA256 TLSv1.2 6000 6000
300 6000 3921.15 br'  ECDHE-ECDSA-AES256-GCM-SHA384 TLSv1.2 6000 6000
-------------------------------------------------------------------------------------------
h2load result summary
min:      avg:      max:      stddev:  requests-succeeded:
2394.000  3391.003  4556.930  572.555  100.00
-------------------------------------------------------------------------------------------
Note these results are testing a single cpu thread only - so more cpu core/threads, the faster the results

eva2000 · Dec 17, 2018

Wordpress 4.9.9 wrk-cmm Benchmarks

I ran benchmarks using my custom forked version of wrk, wrk-cmm. However, the Wordpress 4.9.9 install was created via Centmin Mod 123.09beta01 Wordpress auto installer via centmin.sh menu option 22 with Redis Cache Nginx level caching in place. This means the test never really hit disk or MySQL backend but was mainly done out of caching so was unlikely to be affected by slow disk I/O and MySQL. So looks like newer droplet with Intel Xeon Gold 6140 excelled with more than twice the performance ~137+ % in terms of requests per second !

Also as the test is against HTTPS based Centmin Mod Nginx site, it means cryptographic performance of the cpu is a factor and this is where the newer Intel Xeon Gold 6140 Skylake based cpus would excel.

wrk-cmm install

Code (Text):

git clone -b centminmod https://github.com/centminmod/wrk wrk-cmm
cd wrk-cmm
make
\cp -af wrk /usr/local/bin/wrk-cmm

Code (Text):

wrk-cmm --help
Usage: wrk <options> <url>                  
  Options:                                  
    -c, --connections <N>  Connections to keep open
    -d, --duration    <T>  Duration of test  
    -t, --threads     <N>  Number of threads to use
                                            
    -b, --bind-ip     <S>  Source IP (or CIDR mask)
                                            
    -s, --script      <S>  Load Lua script file
    -H, --header      <H>  Add header to request
        --latency          Print latency statistics
        --breakout         Print breakout statistics
        --timeout     <T>  Socket/request timeout
    -v, --version          Print version details
                                            
  Numeric arguments may include a SI unit (1k, 1M, 1G)
  Time arguments may include a time unit (2s, 2m, 2h)

Using following wrk-cmm test paramters

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com

new 2 cpu droplet Intel Xeon Gold 6140 at 2.30Ghz

5,527.67 requests/s average over 3 runs
14.52ms average TTFB average over 3 runs
65.19ms max TTFB average over 3 runs

old 2 cpu droplet Intel Xeon E5-2630L v2 at 2.40GHz

2,334.07 requests/s average over 3 runs
38.72ms average TTFB average over 3 runs
131.26ms max TTFB average over 3 runs

Raw numbers

new 2 cpu droplet Intel Xeon Gold 6140 at 2.30Ghz

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com
Running 15s test @ https://domain.com
  1 threads and 100 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    17.90ms    8.48ms  57.56ms   73.51%
    Connect    20.24ms   33.63ms  99.35ms    0.00%
    TTFB       14.67ms    8.10ms  54.19ms   71.19%
    TTLB        3.18ms    2.11ms  23.00ms   71.70%
    Req/Sec     5.56k   670.19     6.69k    79.87%
  82921 requests in 15.01s, 2.82GB read
Requests/sec:   5525.40
Transfer/sec:    192.41MB

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com
Running 15s test @ https://domain.com
  1 threads and 100 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    18.24ms    8.79ms  58.19ms   73.68%
    Connect    22.87ms   36.96ms 111.50ms    0.00%
    TTFB       14.87ms    8.41ms  50.12ms   71.56%
    TTLB        3.31ms    2.27ms  26.03ms   69.89%
    Req/Sec     5.47k   554.66     6.32k    72.97%
  81468 requests in 15.03s, 2.77GB read
Requests/sec:   5421.82
Transfer/sec:    188.78MB

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com
Running 15s test @ https://domain.com
  1 threads and 100 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    17.53ms    8.21ms  98.12ms   74.60%
    Connect    69.91ms   37.53ms 153.31ms   39.62%
    TTFB       14.02ms    7.70ms  91.26ms   73.70%
    TTLB        3.45ms    2.39ms  26.17ms   73.59%
    Req/Sec     5.67k   669.61     6.65k    81.88%
  84839 requests in 15.05s, 2.88GB read
Requests/sec:   5635.78
Transfer/sec:    196.23MB

old 2 cpu droplet Intel Xeon E5-2630L v2 at 2.40GHz

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com
Running 15s test @ https://domain.com
  1 threads and 100 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    42.09ms   15.06ms 109.33ms   67.98%
    Connect    51.21ms   77.27ms 215.62ms    0.00%
    TTFB       38.82ms   14.63ms 105.61ms   69.19%
    TTLB        3.27ms    2.09ms  21.12ms   81.25%
    Req/Sec     2.37k   230.45     2.79k    79.59%
  35210 requests in 15.09s, 1.17GB read
Requests/sec:   2332.99
Transfer/sec:     79.18MB

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com
Running 15s test @ https://domain.com
  1 threads and 100 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    42.86ms   15.23ms 116.03ms   66.63%
    Connect   125.20ms   45.80ms 221.95ms   58.87%
    TTFB       39.27ms   14.88ms 106.11ms   67.32%
    TTLB        3.56ms    2.97ms  32.62ms   88.93%
    Req/Sec     2.32k   302.43     2.93k    80.27%
  34434 requests in 15.03s, 1.14GB read
Requests/sec:   2291.52
Transfer/sec:     77.77MB

Code (Text):

wrk-cmm -t1 -c100 -d15s --breakout https://domain.com
Running 15s test @ https://domain.com
  1 threads and 100 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    41.46ms   15.22ms 185.61ms   71.02%
    Connect    52.07ms   60.22ms 157.92ms    0.00%
    TTFB       38.06ms   14.80ms 182.07ms   71.92%
    TTLB        3.37ms    2.51ms  29.48ms   85.32%
    Req/Sec     2.41k   220.59     2.83k    75.51%
  35826 requests in 15.07s, 1.19GB read
Requests/sec:   2377.71
Transfer/sec:     80.70MB

pdinh97qng · Dec 17, 2018

Wow! Thanks for those real numbers! I just thought like you that the new processor should perform much much better than the old one. Looks like the providers now prefers more cores than higher clock speed. If I not wrong that more cores meaning that they can serve more people ) more profit rather than focus on performance of each nodes. If customers want horse power then pay more! I remember those day when Vultr has some kind of 3.x Ghz clock speed CPU and it perform much better these low clock speed newer one

eva2000 · Dec 17, 2018

Yeah as long as you're using a heavily cached stack that offloads load and ensures you don't touch disk I/O and MySQL, then the newer cpus perform better for cpu and memory performance. If anyone else lands on DigitalOcean droplets with Intel Xeon Gold or Platinum cpus, would be interested in your sysbench.sh numbers for cpu, mem, file, file fsync and mysql OLTP as well

eva2000 · Dec 18, 2018

Updated 2nd post with Centmin Mod Nginx HTTP/2 HTTPS benchmarks for both RSA 2048bit and ECDSA 256bit SSL certification configurations too DigitalOcean - Benchmarks - DigitalOcean US$15/month droplet benchmarks - Intel Xeon E5-2630Lv2 vs Intel Xeon Gold 6140

eva2000 · Nov 20, 2021

Roman Jay Almaza said: ↑

What benchmarking softwares do you guys suggests? newbie
Click to expand...

Quick one is the above mentioned/my sysbench script

Log in / Register

Forums

Learn about Centmin Mod LEMP Stack today

DigitalOcean Benchmarks DigitalOcean US$15/month droplet benchmarks - Intel Xeon E5-2630Lv2 vs Intel Xeon Gold 6140

eva2000 Administrator Staff Member

sysbench benchmarks

Conclusion

eva2000 Administrator Staff Member

Nginx HTTP/2 HTTPS Benchmarks

eva2000 Administrator Staff Member

Wordpress 4.9.9 wrk-cmm Benchmarks

pdinh97qng Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

.

Log in / Register

Useful Searches

Learn about Centmin Mod LEMP Stack today

DigitalOcean Benchmarks DigitalOcean US$15/month droplet benchmarks - Intel Xeon E5-2630Lv2 vs Intel Xeon Gold 6140

eva2000 Administrator Staff Member

sysbench benchmarks

Conclusion

eva2000 Administrator Staff Member

Nginx HTTP/2 HTTPS Benchmarks

eva2000 Administrator Staff Member

Wordpress 4.9.9 wrk-cmm Benchmarks

pdinh97qng Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

.