redis压测工具redis-benchmark

redis-benchmark 位于src目录下，是redis自带的压测工具，使用简单方便；

[root@xxxx]# redis-benchmark -p 8200 -c 50 -n 100000   # 50并发，10W的请求量；
====== PING_INLINE ======                       # ping\r\nping\r\nping\r\n.                            100000 requests completed in 1.53 seconds     # 1.53s完成10W的请求50 parallel clients                           # 50并发3 bytes payload                               # 3bytes每次请求keep alive: 1                                 # 单机模式host configuration "save": 3600 1 300 100 60 10000 #save的配置host configuration "appendonly": yes          # AOF模式multi-thread: no                              # 未开启mutl-threadLatency by percentile distribution:  #每次分发一定百分比的延时
0.000% <= 0.087 milliseconds (cumulative count 1)
50.000% <= 0.335 milliseconds (cumulative count 55451) #每次分发一定百分比所对应的延时
75.000% <= 0.543 milliseconds (cumulative count 77491)
87.500% <= 0.559 milliseconds (cumulative count 87688)
93.750% <= 0.575 milliseconds (cumulative count 94537)
96.875% <= 0.599 milliseconds (cumulative count 97281)
98.438% <= 0.631 milliseconds (cumulative count 98528)
99.219% <= 0.671 milliseconds (cumulative count 99242)
99.609% <= 0.767 milliseconds (cumulative count 99627)
99.805% <= 0.975 milliseconds (cumulative count 99805)
99.902% <= 1.647 milliseconds (cumulative count 99903)
99.951% <= 1.703 milliseconds (cumulative count 99954)
99.976% <= 1.879 milliseconds (cumulative count 99976)
99.988% <= 1.999 milliseconds (cumulative count 99988)
99.994% <= 2.055 milliseconds (cumulative count 99994)
99.997% <= 2.095 milliseconds (cumulative count 99998)
99.998% <= 2.111 milliseconds (cumulative count 99999)
99.999% <= 2.119 milliseconds (cumulative count 100000)
100.000% <= 2.119 milliseconds (cumulative count 100000)Cumulative distribution of latencies: #分发一定百分比的累计延时；
0.002% <= 0.103 milliseconds (cumulative count 2)
......
100.000% <= 3.103 milliseconds (cumulative count 100000) #分发完所有请求的总延时Summary:throughput summary: 65189.05 requests per second       #6.5W QPSlatency summary (msec):avg       min       p50       p95       p99       max0.399     0.080     0.335     0.583     0.655     2.119
#平均延时，最小延时，50%、95%、99%的延时，以及最大延时

后续会分别进行SET、GET、INCR等命令的压测；

====== PING_INLINE ======                                                   throughput summary: 65189.05 requests per second
====== PING_MBULK ======                                                   throughput summary: 68493.15 requests per second
====== SET ======                                                   throughput summary: 71326.68 requests per second
====== GET ======                                                   throughput summary: 68775.79 requests per second
====== INCR ======                                                   throughput summary: 71994.23 requests per second
====== LPUSH ======                                                   throughput summary: 65789.48 requests per second
====== RPUSH ======                                                   throughput summary: 65231.57 requests per second
====== LPOP ======                                                   throughput summary: 65919.58 requests per second
====== RPOP ======                                                   throughput summary: 65832.78 requests per second
====== SADD ======                                                   throughput summary: 64516.13 requests per second
====== HSET ======                                                   throughput summary: 69060.77 requests per second
====== SPOP ======                                                   throughput summary: 65316.79 requests per second
====== ZADD ======                                                   throughput summary: 68027.21 requests per second
====== ZPOPMIN ======                                                   throughput summary: 67114.09 requests per second
====== LPUSH (needed to benchmark LRANGE) ======                                                   throughput summary: 71275.84 requests per second
====== LRANGE_100 (first 100 elements) ======                                                   throughput summary: 34317.09 requests per second
====== LRANGE_300 (first 300 elements) ======                                                   throughput summary: 18145.53 requests per second

redis压测工具redis-benchmark相关推荐

Redis压测工具（redis-benchmark）
目录 1.简介 2.参数 3.操作 1.简介 Redis 自带了一个叫 redis-benchmark 的工具来模拟 N 个客户端同时发出 M 个请求.(类似于 Apache ab 程序). Redi ...
Web 压测工具介绍
Web 压测工具 web benchmark tools Apache Bench ApacheBench 是一个用来衡量http服务器性能的单线程命令行工具.原本针对Apache http服务器,但 ...
一个集成redis-benchmark的redis压测的脚本
一个集成redis-benchmark的redis压测的脚本支持单节点主从 sentinel cluster (sentinel的还没写) 出于高可用的考虑,你还可以用高可用的ip做压力测试需 ...
Elasticsearch压测工具esrally使用笔记
Elasticsearch压测工具esrally使用笔记一.环境配置方法一使用docker docker pull elastic/rally docker run elastic/rally ...
精准容量、秒级弹性，压测工具 + SAE 方案如何成功突破传统大促难关？
作者 | 代序阿里云云原生技术团队本文整理自<Serverless 技术公开课>,"Serverless"公众号后台回复"入门",即可获取系列文 ...
从0到1构建美团压测工具
背景美团内部的RPC服务大多构建在Thrift之上,在日常开发服务的过程中,需要针对这些服务进行压力测试(以下简称压测)来发现潜在问题.常用的方法有: 使用一些脚本语言如:Python.Ruby等, ...
golang 压测工具vegeta改造-支持自定义压测任务
背景之前在公司做压测工作的时候,使用了web压测工具vegeta.后续又接到过dns的性能压测.redis的性能压测等任务.http的压测工具vegeta并不能满足需求了.于是模仿vegeta的li ...
HTTP压测工具Wrk 介绍和使用
介绍 wrk是一款简单的HTTP压测工具,托管在Github上,https://github.com/wg/wrk. wrk 的一个很好的特性就是能用很少的线程压出很大的并发量. 原因是它使用了一些操 ...
web版本开源压测工具_Web服务压测神器wrk
wrk是一款开源的高性能http压测工具(也支持https),很是小巧,能够执行文件只有3M(其中主要是luajit和openssl占用绝大多数空间),别看核心代码3-5年没更新了,但依旧很是好用.虽 ...

redis压测工具redis-benchmark

redis压测工具redis-benchmark相关推荐

最新文章

热门文章