sysbench
sysbench是一款开源的多线程性能测试工具,可以执行CPU/内存/线程/IO/数据库等方面的性能测试。数据库目前支持MySQL/Oracle/PostgreSQL,后续的版本oltp测试主要结合了lua脚本,不需要修改源码,通过自定义lua脚本就可以实现不同业务类型的测试;
总结起来sysbench的缺点就是,模拟的表结构太简单,不像tpcc-mysql那样完整的事务系统。但对于性能压测对比还是很有用的,因为sysbench使用的环境参数限制是一样的。
安装
通过git上的源码进行编译安装过程(需要联网)
yum install -y git
yum install -y gcc gcc-c++ automake make libtool
克隆sysbench源码
进入一个目录
git clone https://github.com/akopytov/sysbench.git
cd sysbench
./autogen.sh
#Centos7的mysql-client-libs目录在/var/lib64/mysql目录下
./configure --prefix=/usr/local/sysbench1.0 --with-mysql-includes=/usr/include/mysql --with-mysql-libs=/usr/lib64/mysql
make && make install
/usr/local/sysbench1.0/bin/sysbench --help
[root@server01 sysbench0.5]# ./bin/sysbench --help
Usage:
sysbench [options]... [testname] [command]
Commands implemented by most tests: prepare run cleanup help
General options:
--threads=N number of threads to use [1]
--events=N limit for total number of events [0]
--time=N limit for total execution time in seconds [10]
--warmup-time=N execute events for this many seconds with statistics disabled before the actual benchmark run with statistics enabled [0]
--forced-shutdown=STRING number of seconds to wait after the --time limit before forcing shutdown, or 'off' to disable [off]
--thread-stack-size=SIZE size of stack per thread [64K]
--thread-init-timeout=N wait time in seconds for worker threads to initialize [30]
--rate=N average transactions rate. 0 for unlimited rate [0]
--report-interval=N periodically report intermediate statistics with a specified interval in seconds. 0 disables intermediate reports [0]
--report-checkpoints=[LIST,...] dump full statistics and reset all counters at specified points in time. The argument is a list of comma-separated values representing the amount of time in seconds elapsed from start of test when report checkpoint(s) must be performed. Report checkpoints are off by default. []
--debug[=on|off] print more debugging info [off]
--validate[=on|off] perform validation checks where possible [off]
--help[=on|off] print help and exit [off]
--version[=on|off] print version and exit [off]
--config-file=FILENAME File containing command line options
--luajit-cmd=STRING perform LuaJIT control command. This option is equivalent to 'luajit -j'. See LuaJIT documentation for more information
--tx-rate=N deprecated alias for --rate [0]
--max-requests=N deprecated alias for --events [0]
--max-time=N deprecated alias for --time [0]
--num-threads=N deprecated alias for --threads [1]
Pseudo-Random Numbers Generator options:
--rand-type=STRING random numbers distribution {uniform, gaussian, special, pareto, zipfian} to use by default [special]
--rand-seed=N seed for random number generator. When 0, the current time is used as an RNG seed. [0]
--rand-spec-iter=N number of iterations for the special distribution [12]
--rand-spec-pct=N percentage of the entire range where 'special' values will fall in the special distribution [1]
--rand-spec-res=N percentage of 'special' values to use for the special distribution [75]
--rand-pareto-h=N shape parameter for the Pareto distribution [0.2]
--rand-zipfian-exp=N shape parameter (exponent, theta) for the Zipfian distribution [0.8]
Log options:
--verbosity=N verbosity level {5 - debug, 0 - only critical messages} [3]
--percentile=N percentile to calculate in latency statistics (1-100). Use the special value of 0 to disable percentile calculations [95]
--histogram[=on|off] print latency histogram in report [off]
General database options:
--db-driver=STRING specifies database driver to use ('help' to get list of available drivers)
--db-ps-mode=STRING prepared statements usage mode {auto, disable} [auto]
--db-debug[=on|off] print database-specific debug information [off]
Compiled-in database drivers:
mysql - MySQL driver
mysql options:
--mysql-host=[LIST,...] MySQL server host [localhost]
--mysql-port=[LIST,...] MySQL server port [3306]
--mysql-socket=[LIST,...] MySQL socket
--mysql-user=STRING MySQL user [sbtest]
--mysql-password=STRING MySQL password []
--mysql-db=STRING MySQL database name [sbtest]
--mysql-ssl[=on|off] use SSL connections, if available in the client library [off]
--mysql-ssl-cipher=STRING use specific cipher for SSL connections []
--mysql-compression[=on|off] use compression, if available in the client library [off]
--mysql-debug[=on|off] trace all client library calls [off]
--mysql-ignore-errors=[LIST,...] list of errors to ignore, or "all" [1213,1020,1205]
--mysql-dry-run[=on|off] Dry run, pretend that all MySQL client API calls are successful without executing them [off]
Compiled-in tests:
fileio - File I/O test
cpu - CPU performance test
memory - Memory functions speed test
threads - Threads subsystem performance test
mutex - Mutex performance test
See 'sysbench <testname> help' for a list of options for each test.
测试CPU
Centos7CPU信息查询:
# 总核数 = 物理CPU个数 X 每颗物理CPU的核数
# 总逻辑CPU数 = 物理CPU个数 X 每颗物理CPU的核数 X 超线程数
# 查看物理CPU个数
cat /proc/cpuinfo| grep "physical id"| sort| uniq| wc -l
# 查看每个物理CPU中core的个数(即核数)
cat /proc/cpuinfo| grep "cpu cores"| uniq
# 查看逻辑CPU的个数
cat /proc/cpuinfo| grep "processor"| wc -l
#CPU测试通过计算指定范围内最大素数所需时间
[root@server01 sysbench-0.5]# ./bin/sysbench cpu help
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
cpu options:
--cpu-max-prime=N upper limit for primes generator [10000]
#测试实例,最大素数为5000,线程数等于服务器逻辑CPU个数,预热时间30秒,执行时间60秒,报告输出频率5秒
[root@server01 sysbench-0.5]# ./bin/sysbench cpu --cpu-max-prime=5000 --threads=2 --warmup-time=30 --time=60 --report-interval=5 run
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 2
Warmup time: 30s
Report intermediate results every 5 second(s)
Initializing random number generator from current time
Prime numbers limit: 5000
Initializing worker threads...
Threads started!
Warming up for 30 seconds...
[ 5s ] thds: 2 eps: 5325.41 lat (ms,95%): 0.53
[ 10s ] thds: 2 eps: 5340.05 lat (ms,95%): 0.53
[ 15s ] thds: 2 eps: 5438.63 lat (ms,95%): 0.53
[ 20s ] thds: 2 eps: 5289.38 lat (ms,95%): 0.54
[ 25s ] thds: 2 eps: 5286.74 lat (ms,95%): 0.53
[ 30s ] thds: 2 eps: 5452.40 lat (ms,95%): 0.52
[ 35s ] thds: 2 eps: 5378.61 lat (ms,95%): 0.52
[ 40s ] thds: 2 eps: 5295.81 lat (ms,95%): 0.54
[ 45s ] thds: 2 eps: 5318.48 lat (ms,95%): 0.52
[ 50s ] thds: 2 eps: 5362.93 lat (ms,95%): 0.52
[ 55s ] thds: 2 eps: 5385.05 lat (ms,95%): 0.52
CPU speed:
events per second: 5352.08
Throughput:
events/s (eps): 5352.0840
time elapsed: 60.0013s
total number of events: 321122
Latency (ms): #ms毫秒,us微秒,ns纳秒
min: 0.29
avg: 0.37
max: 36.06
95th percentile: 0.53
sum: 119806.45
Threads fairness:
events (avg/stddev): 160559.0000/1453.00
execution time (avg/stddev): 59.9032/0.00
文件I/O基准测试
#查看操作系统的默认块大小
[root@server01 sysbench-0.5]# stat /boot/|grep "IO Block"
Size: 4096 Blocks: 8 IO Block: 4096 directory
#测试系统在不同I/O负载下的性能
[root@server01 sysbench-0.5]# ./bin/sysbench fileio help
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
fileio options:
--file-num=N number of files to create [128]
--file-block-size=N block size to use in all IO operations [16384] #innodb块16k,OS的默认块大小4k
--file-total-size=SIZE total size of files to create [2G]
--file-test-mode=STRING test mode {seqwr顺序写, seqrewr顺序读写, seqrd顺序读, rndrd随机读, rndwr随机写, rndrw随机读写}
--file-io-mode=STRING file operations mode {sync同步,async异步,mmap} [sync]
--file-extra-flags=[LIST,...] list of additional flags to use to open files {sync,dsync,direct} []
--file-fsync-freq=N do fsync() after this number of requests (0 - don't use fsync()) [100]
--file-fsync-all[=on|off] do fsync() after each write operation [off]
--file-fsync-end[=on|off] do fsync() at the end of test [on]
--file-fsync-mode=STRING which method to use for synchronization {fsync, fdatasync} [fsync]
--file-merged-requests=N merge at most this number of IO requests if possible (0 - don't merge) [0]
--file-rw-ratio=N reads/writes ratio for combined test [1.5]
#准备阶段
生成测试用的数据文件,要求生成的数据文件至少要比内存大,否则数据由于被操作系统缓存而无法体现 I/O 密集型工作负载。
[root@server01 sysbench-0.5]# ./bin/sysbench fileio --file-total-size=5G --file-test-mode=rndrw --file-block-size=16384 prepare
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
128 files, 40960Kb each, 5120Mb total
Creating files for the test...
Extra file open flags: (none)
Creating file test_file.0
......
Creating file test_file.127
5368709120 bytes written in 345.57 seconds (14.82 MiB/sec).
[root@server01 sysbench-0.5]# ls -lh
total 5.0G
drwxr-xr-x. 2 root root 21 Mar 16 17:57 bin
drwxr-xr-x. 3 root root 21 Mar 16 17:57 share
-rw-------. 1 root root 40M Mar 19 10:18 test_file.0
#运行阶段,利用前一步生成的文件,运行测试
[root@server01 sysbench-0.5]# ./bin/sysbench fileio --file-total-size=5G --file-test-mode=rndrw --file-block-size=16384 --threads=2 --warmup-time=30 --time=60 --report-interval=5 run
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 2
Warmup time: 30s
Report intermediate results every 5 second(s)
Initializing random number generator from current time
Extra file open flags: (none)
128 files, 40MiB each
5GiB total file size
Block size 16KiB
Number of IO requests: 0
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
Initializing worker threads...
Threads started!
Warming up for 30 seconds...
[ 5s ] reads: 0.73 MiB/s writes: 0.49 MiB/s fsyncs: 98.73/s latency (ms,95%): 58.923
[ 10s ] reads: 0.73 MiB/s writes: 0.48 MiB/s fsyncs: 102.40/s latency (ms,95%): 38.942
[ 15s ] reads: 0.92 MiB/s writes: 0.61 MiB/s fsyncs: 127.00/s latency (ms,95%): 34.330
[ 20s ] reads: 1.12 MiB/s writes: 0.75 MiB/s fsyncs: 129.62/s latency (ms,95%): 34.330
[ 25s ] reads: 1.13 MiB/s writes: 0.75 MiB/s fsyncs: 178.58/s latency (ms,95%): 34.954
[ 30s ] reads: 0.95 MiB/s writes: 0.63 MiB/s fsyncs: 128.00/s latency (ms,95%): 33.116
[ 35s ] reads: 1.08 MiB/s writes: 0.72 MiB/s fsyncs: 127.99/s latency (ms,95%): 33.718
[ 40s ] reads: 0.83 MiB/s writes: 0.56 MiB/s fsyncs: 128.00/s latency (ms,95%): 32.525
[ 45s ] reads: 0.91 MiB/s writes: 0.61 MiB/s fsyncs: 128.01/s latency (ms,95%): 35.589
[ 50s ] reads: 0.77 MiB/s writes: 0.52 MiB/s fsyncs: 102.40/s latency (ms,95%): 41.851
[ 55s ] reads: 0.85 MiB/s writes: 0.57 MiB/s fsyncs: 102.40/s latency (ms,95%): 32.525
[ 60s ] reads: 0.98 MiB/s writes: 0.65 MiB/s fsyncs: 139.60/s latency (ms,95%): 35.589
Throughput:
read: IOPS=57.12 0.89 MiB/s (0.94 MB/s)
write: IOPS=38.06 0.59 MiB/s (0.62 MB/s)
fsync: IOPS=122.58
Latency (ms):
min: 0.00
avg: 9.19
max: 948.38
95th percentile: 36.24
sum: 120152.38
#清除阶段
删除测试过程中的文件
[root@server01 sysbench-0.5]# ./bin/sysbench fileio --file-total-size=5G cleanup sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Removing test files...
[root@server01 sysbench-0.5]# ls -lh
total 0
drwxr-xr-x. 2 root root 21 Mar 16 17:57 bin
drwxr-xr-x. 3 root root 21 Mar 16 17:57 share
内存测试
[root@server01 sysbench-0.5]# ./bin/sysbench memory help
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
memory options:
--memory-block-size=SIZE size of memory block for test [1K]
--memory-total-size=SIZE total size of data to transfer [100G]
--memory-scope=STRING memory access scope {global,local} [global]
--memory-hugetlb[=on|off] allocate memory from HugeTLB pool [off]
--memory-oper=STRING type of memory operations {read, write, none} [write]
--memory-access-mode=STRING memory access mode {seq,rnd} [seq]
[root@server01 sysbench-0.5]# ./bin/sysbench memory --memory-block-size=4k --memory-total-size=1G --memory-access-mode=rnd run
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 4KiB
total size: 1024MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 262144 (598832.43 per second)
1024.00 MiB transferred (2339.19 MiB/sec)
Throughput:
events/s (eps): 598832.4339
time elapsed: 0.4378s
total number of events: 262144
Latency (ms):
min: 0.00
avg: 0.00
max: 0.17
95th percentile: 0.00
sum: 396.48
Threads fairness:
events (avg/stddev): 262144.0000/0.00
execution time (avg/stddev): 0.3965/0.00
线程测试
[root@server01 sysbench-0.5]# ./bin/sysbench threads help
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
threads options:
--thread-yields=N number of yields to do per request [1000] #请求产生的线程
--thread-locks=N number of locks per thread [8] #每个线程的锁
[root@server01 sysbench-0.5]# ./bin/sysbench threads --threads=2 --time=60 --warmup-time=30 run
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 2
Warmup time: 30s
Initializing random number generator from current time
Initializing worker threads...
Threads started!
Warming up for 30 seconds...
Throughput:
events/s (eps): 5041.7009
time elapsed: 60.0002s
total number of events: 302499
Latency (ms):
min: 0.28
avg: 0.40
max: 18.76
95th percentile: 0.77
sum: 119842.00
Threads fairness:
events (avg/stddev): 151247.5000/30.50
execution time (avg/stddev): 59.9210/0.00
OLTP测试
对于mysql的OLTP测试,和file一样,同样需要经历prepare,run,cleanup三个阶段。prepare阶段会在数据库中指定张和指定行数的表,默认表在sbtest架构下,表名为sbtestN,数据库sbtest需要手动在数据库中创建否则会报错。
通过调用不同的lua脚本测试不同业务模式下的数据库性能,也可自定义lua脚本进行测试。
[root@server01 sysbench0.5]# ./bin/sysbench share/sysbench/oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-user=root --mysql-password='Aa123456789' --mysql-socket=/data/mysqldata3306/sock/mysql.sock --mysql-db=sbtest --db-driver=mysql --tables=10 --table-size=1000000 --report-interval=10 --threads=4 --time=120 prepare
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Initializing worker threads...
Creating table 'sbtest3'...
......
Creating a secondary index on 'sbtest10'...
[root@server01 sysbench0.5]# ./bin/sysbench share/sysbench/oltp_read_write.lua --mysql-host=localhost --mysql-port=3306 --mysql-user=root --mysql-password='Aa123456789' --mysql-socket=/data/mysqldata3306/sock/mysql.sock --mysql-db=sbtest --db-driver=mysql --tables=10 --table-size=1000000 --report-interval=10 --threads=4 --time=120 run
sysbench 1.1.0-651e7fd (using bundled LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Report intermediate results every 10 second(s)
Initializing random number generator from current time
......
SQL statistics:
queries performed:
read: 4340
write: 1240
other: 620
total: 6200
transactions: 310 (2.55 per sec.) #TPS
queries: 6200 (51.05 per sec.) #QPS
ignored errors: 0 (0.00 per sec.)
reconnects: 0 (0.00 per sec.)
Throughput:
events/s (eps): 2.5523
time elapsed: 121.4602s
total number of events: 310
Latency (ms):
min: 312.48
avg: 1564.91 #平均响应时间
max: 3674.74
95th percentile: 3040.14 #95%执行响应时间
sum: 485121.69
Threads fairness:
events (avg/stddev): 77.5000/1.12
execution time (avg/stddev): 121.2804/0.14