Redis-benchmark是官方自带的Redis性能测试工具，可以有效的测试Redis服务的性能。
使用说明如下:
1 Usage: redis-benchmark [-h <host>] [-p <port>] [-c <clients>] [-n <requests]> [-k <boolean>] 2 3 -h <hostname> Server hostname (default 127.0.0.1) 4 -p <port> Server port (default 6379) 5 -s <socket> Server socket (overrides host and port) 6 -c <clients> Number of parallel connections (default 50) 7 -n <requests> Total number of requests (default 10000) 8 -d <size> Data size of SET/GET value in bytes (default 2) 9 -k <boolean> 1=keep alive 0=reconnect (default 1) 10 -r <keyspacelen> Use random keys for SET/GET/INCR, random values for SADD 11 Using this option the benchmark will get/set keys 12 in the form mykey_rand:000000012456 instead of constant 13 keys, the <keyspacelen> argument determines the max 14 number of values for the random number. For instance 15 if set to 10 only rand:000000000000 - rand:000000000009 16 range will be allowed. 17 -P <numreq> Pipeline <numreq> requests. Default 1 (no pipeline). 18 -q Quiet. Just show query/sec values 19 --csv Output in CSV format 20 -l Loop. Run the tests forever 21 -t <tests> Only run the comma-separated list of tests. The test 22 names are the same as the ones produced as output. 23 -I Idle mode. Just open N idle connections and wait.
测试命令事例
1、redis-benchmark -h 127.0.0.1 -p 6379 -c 100 -n 100000
100个并发连接，100000个请求，检测host为localhost 端口为6379的redis服务器性能
2、redis-benchmark -h 127.0.0.1 -p 6379 -q -d 100
测试存取大小为100字节的数据包的性能
3、redis-benchmark -t set,lpush -n 100000 -q
只测试某些操作的性能
4、redis-benchmark -n 100000 -q script load "redis.call('set','foo','bar')"
只测试某些数值存取的性能
测试结果分析
1 [root@H__D redis]# redis-benchmark -h 127.0.0.1 -p 6379 -c 100 -n 100000
2 ====== PING_INLINE ======
3 100000 requests completed in 0.94 seconds
4 100 parallel clients
5 3 bytes payload
6 keep alive: 1
7
8 99.50% <= 1 milliseconds
9 99.97% <= 2 milliseconds
10 100.00% <= 2 milliseconds
11 106951.88 requests per second
12
13 ====== PING_BULK ======
14 100000 requests completed in 0.89 seconds
15 100 parallel clients
16 3 bytes payload
17 keep alive: 1
18
19 99.43% <= 1 milliseconds
20 99.98% <= 2 milliseconds
21 100.00% <= 2 milliseconds
22 111856.82 requests per second
23
24 ====== SET ======
25 100000 requests completed in 0.92 seconds
26 100 parallel clients
27 3 bytes payload
28 keep alive: 1
29
30 99.67% <= 1 milliseconds
31 100.00% <= 2 milliseconds
32 108459.87 requests per second
33
34 ====== GET ======
35 100000 requests completed in 0.91 seconds
36 100 parallel clients
37 3 bytes payload
38 keep alive: 1
39
40 99.60% <= 1 milliseconds
41 100.00% <= 2 milliseconds
42 100.00% <= 2 milliseconds
43 109769.48 requests per second
44
45 ====== INCR ======
46 100000 requests completed in 0.92 seconds
47 100 parallel clients
48 3 bytes payload
49 keep alive: 1
50
51 99.64% <= 1 milliseconds
52 100.00% <= 2 milliseconds
53 100.00% <= 2 milliseconds
54 109289.62 requests per second
55
56 ====== LPUSH ======
57 100000 requests completed in 0.92 seconds
58 100 parallel clients
59 3 bytes payload
60 keep alive: 1
61
62 99.71% <= 1 milliseconds
63 100.00% <= 2 milliseconds
64 100.00% <= 2 milliseconds
65 109170.30 requests per second
66
67 ====== RPUSH ======
68 100000 requests completed in 0.92 seconds
69 100 parallel clients
70 3 bytes payload
71 keep alive: 1
72
73 99.64% <= 1 milliseconds
74 99.99% <= 2 milliseconds
75 100.00% <= 2 milliseconds
76 108695.65 requests per second
77
78 ====== LPOP ======
79 100000 requests completed in 0.91 seconds
80 100 parallel clients
81 3 bytes payload
82 keep alive: 1
83
84 99.62% <= 1 milliseconds
85 99.99% <= 2 milliseconds
86 100.00% <= 2 milliseconds
87 109529.02 requests per second
88
89 ====== RPOP ======
90 100000 requests completed in 0.91 seconds
91 100 parallel clients
92 3 bytes payload
93 keep alive: 1
94
95 99.74% <= 1 milliseconds
96 99.99% <= 2 milliseconds
97 100.00% <= 2 milliseconds
98 109409.20 requests per second
99
100 ====== SADD ======
101 100000 requests completed in 0.92 seconds
102 100 parallel clients
103 3 bytes payload
104 keep alive: 1
105
106 99.61% <= 1 milliseconds
107 99.99% <= 2 milliseconds
108 100.00% <= 2 milliseconds
109 109051.26 requests per second
110
111 ====== HSET ======
112 100000 requests completed in 0.93 seconds
113 100 parallel clients
114 3 bytes payload
115 keep alive: 1
116
117 99.32% <= 1 milliseconds
118 99.98% <= 2 milliseconds
119 100.00% <= 2 milliseconds
120 107642.62 requests per second
121
122 ====== SPOP ======
123 100000 requests completed in 0.90 seconds
124 100 parallel clients
125 3 bytes payload
126 keep alive: 1
127
128 99.76% <= 1 milliseconds
129 100.00% <= 1 milliseconds
130 110741.97 requests per second
131
132 ====== LPUSH (needed to benchmark LRANGE) ======
133 100000 requests completed in 0.92 seconds
134 100 parallel clients
135 3 bytes payload
136 keep alive: 1
137
138 99.69% <= 1 milliseconds
139 99.99% <= 2 milliseconds
140 100.00% <= 2 milliseconds
141 108577.63 requests per second
142
143 ====== LRANGE_100 (first 100 elements) ======
144 100000 requests completed in 0.93 seconds
145 100 parallel clients
146 3 bytes payload
147 keep alive: 1
148
149 99.77% <= 1 milliseconds
150 100.00% <= 1 milliseconds
151 108108.11 requests per second
152
153 ====== LRANGE_300 (first 300 elements) ======
154 100000 requests completed in 0.93 seconds
155 100 parallel clients
156 3 bytes payload
157 keep alive: 1
158
159 99.52% <= 1 milliseconds
160 100.00% <= 2 milliseconds
161 100.00% <= 2 milliseconds
162 107296.14 requests per second
163
164 ====== LRANGE_500 (first 450 elements) ======
165 100000 requests completed in 0.94 seconds
166 100 parallel clients
167 3 bytes payload
168 keep alive: 1
169
170 99.64% <= 1 milliseconds
171 99.99% <= 2 milliseconds
172 100.00% <= 2 milliseconds
173 106723.59 requests per second
174
175 ====== LRANGE_600 (first 600 elements) ======
176 100000 requests completed in 0.92 seconds
177 100 parallel clients
178 3 bytes payload
179 keep alive: 1
180
181 99.73% <= 1 milliseconds
182 100.00% <= 2 milliseconds
183 108225.10 requests per second
184
185 ====== MSET (10 keys) ======
186 100000 requests completed in 1.06 seconds
187 100 parallel clients
188 3 bytes payload
189 keep alive: 1
190
191 91.36% <= 1 milliseconds
192 99.97% <= 2 milliseconds
193 100.00% <= 2 milliseconds
194 94517.96 requests per second
1 [root@H__D redis]# redis-benchmark -h 127.0.0.1 -p 6379 -q -d 100 2 PING_INLINE: 113122.17 requests per second 3 PING_BULK: 117785.63 requests per second 4 SET: 114285.71 requests per second 5 GET: 116144.02 requests per second 6 INCR: 116144.02 requests per second 7 LPUSH: 113895.21 requests per second 8 RPUSH: 113507.38 requests per second 9 LPOP: 116550.12 requests per second 10 RPOP: 116279.07 requests per second 11 SADD: 115740.73 requests per second 12 HSET: 112866.82 requests per second 13 SPOP: 116279.07 requests per second 14 LPUSH (needed to benchmark LRANGE): 113636.37 requests per second 15 LRANGE_100 (first 100 elements): 114025.09 requests per second 16 LRANGE_300 (first 300 elements): 113895.21 requests per second 17 LRANGE_500 (first 450 elements): 113507.38 requests per second 18 LRANGE_600 (first 600 elements): 114155.25 requests per second 19 MSET (10 keys): 93023.25 requests per second
