redis性能压测

---

redis性能压测

redis性能压测工具–redis-benchmark

如果要对 redis 进行性能压测，可以选择 redis 自带的压测工具 redis-benchmark。

进入到 /usr/local/software/redis-6.0.9/src 下，使用如下命令开始 redis 性能压测：

./redis-benchmark -h 10.0.0.4 -p 6379 -a redis-pass -c 5000 -n 50000 -d 50

使用 redis-benchmark -h 命令可以查看 redis-benchmark 相关参数，如下所示：

Usage: redis-benchmark [-h <host>] [-p <port>] [-c <clients>] [-n <requests>] [-k <boolean>]

 -h <hostname>      Server hostname (default 127.0.0.1)
 -p <port>          Server port (default 6379)
 -s <socket>        Server socket (overrides host and port)
 -a <password>      Password for Redis Auth
 --user <username>  Used to send ACL style 'AUTH username pass'. Needs -a.
 -c <clients>       Number of parallel connections (default 50)
 -n <requests>      Total number of requests (default 100000)
 -d <size>          Data size of SET/GET value in bytes (default 3)
 --dbnum <db>       SELECT the specified db number (default 0)
 --threads <num>    Enable multi-thread mode.
 --cluster          Enable cluster mode.
 --enable-tracking  Send CLIENT TRACKING on before starting benchmark.
 -k <boolean>       1=keep alive 0=reconnect (default 1)
 -r <keyspacelen>   Use random keys for SET/GET/INCR, random values for SADD,
                    random members and scores for ZADD.
  Using this option the benchmark will expand the string __rand_int__
  inside an argument with a 12 digits number in the specified range
  from 0 to keyspacelen-1. The substitution changes every time a command
  is executed. Default tests use this to hit random keys in the
  specified range.
 -P <numreq>        Pipeline <numreq> requests. Default 1 (no pipeline).
 -e                 If server replies with errors, show them on stdout.
                    (no more than 1 error per second is displayed)
 -q                 Quiet. Just show query/sec values
 --precision        Number of decimal places to display in latency output (default 0)
 --csv              Output in CSV format
 -l                 Loop. Run the tests forever
 -t <tests>         Only run the comma separated list of tests. The test
                    names are the same as the ones produced as output.
 -I                 Idle mode. Just open N idle connections and wait.

redis-benchmark 相关参数解释说明

redis-benchmark 压测结果

redis-benchmark 压测结果如下所示：

[root@mq1 src]# ./redis-benchmark -h 10.0.0.4 -p 6379 -a redis-pass -c 5000 -n 50000 -d 50
====== PING_INLINE ======
  50000 requests completed in 0.74 seconds
  5000 parallel clients
  50 bytes payload
  keep alive: 1
  host configuration "save": 900 1 300 10 60 10000
  host configuration "appendonly": yes
  multi-thread: no

0.00% <= 18 milliseconds
0.11% <= 19 milliseconds
0.30% <= 20 milliseconds
0.78% <= 21 milliseconds
1.13% <= 22 milliseconds
1.36% <= 23 milliseconds
1.71% <= 24 milliseconds
2.31% <= 25 milliseconds
3.44% <= 26 milliseconds
5.26% <= 27 milliseconds
7.53% <= 28 milliseconds
9.69% <= 29 milliseconds
11.74% <= 30 milliseconds
13.81% <= 31 milliseconds
15.82% <= 32 milliseconds
17.74% <= 33 milliseconds
20.13% <= 34 milliseconds
22.83% <= 35 milliseconds
25.27% <= 36 milliseconds
27.70% <= 37 milliseconds
30.22% <= 38 milliseconds
32.75% <= 39 milliseconds
35.68% <= 40 milliseconds
39.77% <= 41 milliseconds
45.01% <= 42 milliseconds
50.80% <= 43 milliseconds
55.11% <= 44 milliseconds
58.44% <= 45 milliseconds
61.85% <= 46 milliseconds
64.59% <= 47 milliseconds
67.36% <= 48 milliseconds
69.99% <= 49 milliseconds
72.79% <= 50 milliseconds
75.65% <= 51 milliseconds
78.24% <= 52 milliseconds
80.39% <= 53 milliseconds
82.34% <= 54 milliseconds
84.34% <= 55 milliseconds
86.57% <= 56 milliseconds
87.80% <= 57 milliseconds
88.93% <= 58 milliseconds
89.99% <= 59 milliseconds
90.70% <= 60 milliseconds
92.08% <= 61 milliseconds
92.78% <= 62 milliseconds
93.35% <= 63 milliseconds
93.93% <= 64 milliseconds
94.44% <= 65 milliseconds
94.70% <= 66 milliseconds
95.02% <= 67 milliseconds
95.29% <= 68 milliseconds
95.53% <= 69 milliseconds
95.73% <= 70 milliseconds
96.04% <= 71 milliseconds
96.34% <= 72 milliseconds
96.64% <= 73 milliseconds
97.06% <= 74 milliseconds
97.44% <= 75 milliseconds
97.78% <= 76 milliseconds
97.90% <= 77 milliseconds
98.17% <= 78 milliseconds
98.32% <= 79 milliseconds
98.49% <= 80 milliseconds
98.71% <= 81 milliseconds
98.78% <= 82 milliseconds
98.97% <= 83 milliseconds
99.12% <= 84 milliseconds
99.26% <= 85 milliseconds
99.41% <= 86 milliseconds
99.57% <= 87 milliseconds
99.72% <= 88 milliseconds
99.86% <= 89 milliseconds
100.00% <= 89 milliseconds
67659.00 requests per second

......

redis-benchmark 压测踩坑

踩坑一：Can’t create socket: Too many open files

问题描述：

linux 系统默认 open file 是 1024，程序打开的文件 /socket 连接数量超过系统设定值。可以通过 ulimit -a 命令查看，如下所示：

解决方案：

1. 通过 ulimit -n 命令临时扩大 open file，如下所示：

ulimit -n 40960

2. 修改 /etc/security/limits.conf 文件永久生效扩大 open file，在文件最后加上以下代码，如下所示：

* soft nofile 40960
* hard nofile 40960

踩坑二：Cannot assign requested address

问题描述：

这是由于 redis-benchmark 的压力测试时频繁的连服务器，数据量较大的时候，因为每次链接都在很短的时间内结束，致使不少的 TIME_WAIT，以致于用光了可用的端口号，因此新的链接没办法绑定端口，即 “Cannot assign requestedaddress”。

解决方案：

执行以下命令：

# 开启对于TCP时间戳的支持,若该项设置为0，则下面一项设置不起作用
sysctl -w net.ipv4.tcp_timestamps=1 
# 表示开启TCP连接中TIME-WAIT sockets的快速回收
sysctl -w net.ipv4.tcp_tw_recycle=1 

然而，执行上述命令后，再次使上述命令进行压测，仍然出现上述问题。因为上述命令设置的请求数和连接数比较大，所以尝试性的减小了其数值，执行之后成功开启 redis-benchmark 的压力测试，如下所示：