# 先确定cpu是几核的,然后在设置后续的测试使用的线程数为该数值的2倍, 但是这里测试的是单块硬盘,应当以单线程进行测试。
cat /proc/cpuinfo | grep "model name" | wc -l
lsscsi -c # 查看硬盘信息
fdisk -l # 查看分区信息
df -h # 参看分区使用状况
# 在当前目录创建所需的测试文件
sysbench --test=fileio --num-threads=8 --file-total-size=3G --file-test-mode=rndrw prepare
# 运行测试
sysbench --test=fileio --num-threads=8 --file-total-size=3G --file-test-mode=rndrw run
# 清理
sysbench --test=fileio --num-threads=8 --file-total-size=3G --file-test-mode=rndrw cleanup
target | rndrd.transfer | rndrd.iops | rndwr.transfer | rndwr.iops | rndrw.transfer | rndrw.iops |
---|---|---|---|---|---|---|
zll's pc | 3.8934Mb/sec | 249.18 req/sec | 1.0622Mb/sec | 67.98 req/sec | 1.7472Mb/sec | 111.82 req/sec |
sever ssd | 4.6985Gb/sec | 307921.13 req/sec | 244.37Mb/sec | 15639.80 req/sec | 540.66Mb/sec | 34602.23 req/sec |
server | 3.5748Gb/sec | 234278.68 req/sec | 12.246Mb/sec | 783.75 req/sec | 23.766Mb/sec | 1521.04 req/sec |
Aliyun | 3.5868Gb/sec | 235062.80 req/sec | 6.7517Mb/sec | 432.11 req/sec | 9.1607Mb/sec | 586.29 req/sec |
zll普通台式机硬盘
cpu: Intel(R) Core(TM) i3-3240 CPU @ 3.40GHz * 4
硬盘 : ATA WDC WD10EZEX-08M 1TB 7200 RPM
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 8
Extra file open flags: 0
128 files, 24Mb each
3Gb total file size
Block size 16Kb
Number of random requests for random IO: 10000
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Threads started!
Done.
Operations performed: 6009 Read, 4002 Write, 12800 Other = 22811 Total
Read 93.891Mb Written 62.531Mb Total transferred 156.42Mb (1.7472Mb/sec)
111.82 Requests/sec executed
Test execution summary:
total time: 89.5257s
total number of events: 10011
total time taken by event execution: 200.3244
per-request statistics:
min: 0.00ms
avg: 20.01ms
max: 441.67ms
approx. 95 percentile: 96.36ms
Threads fairness:
events (avg/stddev): 1251.3750/49.66
execution time (avg/stddev): 25.0405/0.95
服务器90硬盘 SSD?
CPU: Intel(R) Xeon(R) CPU E3-1230 V2 @ 3.30GHz * 8
硬盘: ATA Samsung SSD 840
sysbench 0.4.12: multi-threaded system evaluation benchmark
128 files, 24576Kb each, 3072Mb total
Creating files for the test...
[root@prod90 1]# sysbench --test=fileio --num-threads=16 --file-total-size=3G --file-test-mode=rndrw run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 16
Extra file open flags: 0
128 files, 24Mb each
3Gb total file size
Block size 16Kb
Number of random requests for random IO: 10000
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Threads started!
Done.
Operations performed: 6004 Read, 4003 Write, 12803 Other = 22810 Total
Read 93.812Mb Written 62.547Mb Total transferred 156.36Mb (540.66Mb/sec)
34602.23 Requests/sec executed
Test execution summary:
total time: 0.2892s
total number of events: 10007
total time taken by event execution: 0.3881
per-request statistics:
min: 0.00ms
avg: 0.04ms
max: 24.90ms
approx. 95 percentile: 0.04ms
Threads fairness:
events (avg/stddev): 625.4375/165.06
execution time (avg/stddev): 0.0243/0.01
服务器80硬盘
CPU : Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz * 24
硬盘: SEAGATE ST3300657SS 15000转/分
[root@lizi80 1]# cat /proc/cpuinfo | grep "model name" | wc -l
24
[root@lizi80 1]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda2 625G 26G 568G 5% /
tmpfs 16G 0 16G 0% /dev/shm
/dev/sda1 2.0G 75M 1.8G 4% /boot
/dev/sda5 97G 18G 74G 19% /data0
/dev/sda6 53G 180M 51G 1% /data1
/dev/sda3 193G 27G 156G 15% /home
/dev/sda7 29G 172M 28G 1% /opt
/dev/sda11 9.7G 169M 9.0G 2% /tmp
/dev/sda8 29G 1.3G 27G 5% /usr
/dev/sda9 29G 516M 27G 2% /usr/local
/dev/sda10 29G 320M 28G 2% /var
192.168.71.20:/data0/web-app/b2c/user_upload
533G 350G 156G 70% /home/www/user_upload
[root@lizi80 1]# pwd
/data0/tmp/1
/data0/tmp/1 测试结果
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 48
Extra file open flags: 0
128 files, 24Mb each
3Gb total file size
Block size 16Kb
Number of random requests for random IO: 10000
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Threads started!
Done.
Operations performed: 6044 Read, 4032 Write, 12765 Other = 22841 Total
Read 94.438Mb Written 63Mb Total transferred 157.44Mb (23.766Mb/sec)
1521.04 Requests/sec executed
Test execution summary:
total time: 6.6244s
total number of events: 10076
total time taken by event execution: 20.8872
per-request statistics:
min: 0.00ms
avg: 2.07ms
max: 186.28ms
approx. 95 percentile: 9.78ms
Threads fairness:
events (avg/stddev): 209.9167/33.04
execution time (avg/stddev): 0.4352/0.11