Redis压测工具redisbenchmark

云计算 2023-08-15 贤蛋大眼萌手机阅读

redis-benchmark 压测参数

参数	参数说明
-h	redis 连接地址
-p	redis端口号，不写默认为6379，tls默认为6479
-a	redis 实例的用户账号密码，无密码时可不写。如果是default账号，直接写`<paasword>`即可。如果是新创建的账号，则填写`<user>:<password>`
-n	测试的总请求数，可设置较大的值以持续压测
-r	随机使用key数量，即使用多少个不同Key
-c	并发的客户端连接数
-d	SET或GEt所操作的值的结果大小，单位为字节(Byte)
-t	需要执行的测试命令，如SET、GET
`--threads`	启动多线程压测，并指定线程个数
`--cluster`	启动集群模式压测，测试整个redis cluster
--tls	启用安全TLS连接
--cert	用于进行身份验证的客户端证书
--key	用于进行身份验证的私钥文件
--cacert	用于验证的CA证书文件

redis-benchmark 测试指标

指标	说明
QPS	表示每秒处理的操作数，单位为次(秒)

安装redis-benchmark工具

<span>[</span>
<span>root@k8s</span>
<span>-</span>
<span>02</span>
<span> opt</span>
<span>]#</span>
<span> wget https</span>
<span>:</span>
<span>//download.redis.io/releases/redis-6.2.5.tar.gz</span>
<span>[</span>
<span>root@k8s</span>
<span>-</span>
<span>02</span>
<span> redis</span>
<span>-</span>
<span>6.2</span>
<span>.</span>
<span>5</span>
<span>]#</span>
<span> tar xf redis</span>
<span>-</span>
<span>6.2</span>
<span>.</span>
<span>5.tar</span>
<span>.</span>
<span>gz </span>
<span>&&</span>
<span> cd redis</span>
<span>-</span>
<span>6.2</span>
<span>.</span>
<span>5</span>
<span>[</span>
<span>root@k8s</span>
<span>-</span>
<span>02</span>
<span> redis</span>
<span>-</span>
<span>6.2</span>
<span>.</span>
<span>5</span>
<span>]#</span>
<span> make </span>
<span>&&</span>
<span> make install</span>

安装完成后在src目录下生成redis-benchmark二进制

redis-benchmark 常见压测案例

这里我们已经创建好redis-cluster，我们在k8s节点上直接执行,进行压测工作

<span>[</span>
<span>root@k8s</span>
<span>-</span>
<span>01</span>
<span> </span>
<span>~]#</span>
<span> kubectl </span>
<span>get</span>
<span> svc </span>
<span>-</span>
<span>n redis</span>
<span>NAME                            TYPE        CLUSTER</span>
<span>-</span>
<span>IP      EXTERNAL</span>
<span>-</span>
<span>IP   PORT</span>
<span>(</span>
<span>S</span>
<span>)</span>
<span>              AGE</span>
<span>redis</span>
<span>-</span>
<span>cluster                   </span>
<span>ClusterIP</span>
<span>   </span>
<span>10.96</span>
<span>.</span>
<span>64.112</span>
<span>    </span>
<span><none></span>
<span>        </span>
<span>6379</span>
<span>/</span>
<span>TCP             </span>
<span>49d</span>

具体压测数据 10GB测试数据，set 1024字节数据，启动32个线程、512连接、100000个key的取值范围

<span>[</span>
<span>root@k8s</span>
<span>-</span>
<span>node2 redis</span>
<span>-</span>
<span>6.2</span>
<span>.</span>
<span>5</span>
<span>]#</span>
<span> </span>
<span>./</span>
<span>src</span>
<span>/</span>
<span>redis</span>
<span>-</span>
<span>benchmark </span>
<span>-</span>
<span>h </span>
<span>10.96</span>
<span>.</span>
<span>64.112</span>
<span>  </span>
<span>-</span>
<span>p </span>
<span>6379</span>
<span> </span>
<span>-</span>
<span>n </span>
<span>10000000</span>
<span> </span>
<span>-</span>
<span>r </span>
<span>100000</span>
<span>  </span>
<span>-</span>
<span>c </span>
<span>512</span>
<span> </span>
<span>-</span>
<span>t </span>
<span>set</span>
<span> </span>
<span>-</span>
<span>d  </span>
<span>1024</span>
<span> </span>
<span>--</span>
<span>threads </span>
<span>32</span>
<span> </span>
<span>--</span>
<span>cluster  </span>
<span>-</span>
<span>a redis123</span>
<span>Cluster</span>
<span> has </span>
<span>3</span>
<span> master nodes</span>
<span>:</span>

<span>Master</span>
<span> </span>
<span>0</span>
<span>:</span>
<span> </span>
<span>35af7e1463aec91be23dec0266682ac0b52da262</span>
<span> </span>
<span>10.244</span>
<span>.</span>
<span>1.12</span>
<span>:</span>
<span>6379</span>
<span>Master</span>
<span> </span>
<span>1</span>
<span>:</span>
<span> f5039b4035bbc66f0562efdd2e91bc4a28064d4f </span>
<span>10.244</span>
<span>.</span>
<span>1.10</span>
<span>:</span>
<span>6379</span>
<span>Master</span>
<span> </span>
<span>2</span>
<span>:</span>
<span> </span>
<span>14813ef24b9f5ca34151ed6eaad21af1f885d6c9</span>
<span> </span>
<span>10.244</span>
<span>.</span>
<span>2.109</span>
<span>:</span>
<span>6379</span>

<span>SET</span>
<span>:</span>
<span> rps</span>
<span>=</span>
<span>90000.0</span>
<span> </span>
<span>(</span>
<span>overall</span>
<span>:</span>
<span> </span>
<span>36413.7</span>
<span>)</span>
<span> avg_msec</span>
<span>=</span>
<span>7.698</span>
<span> </span>
<span>(</span>
<span>overall</span>
<span>:</span>
<span> </span>
<span>13.670</span>
<span>))</span>
<span>6</span>
<span>))</span>

<span>...</span>
<span>#因为我这里是E5服务器，我们等待压测完毕即可</span>
<span>#上面的相关参数可以参考上面的参数表，这里不再过多解释</span>