Redis性能测试：不同value大小对QPS和响应时间的影响-优快云博客

本文链接：https://blog.youkuaiyun.com/qian_348840260/article/details/110670688

在单机Redis 5.0.7环境下，使用4G内存和1000M带宽进行测试。通过设置不同value大小（20000、2000、200、20字节），在200客户端并发下发送100000请求。测试结果显示：string类型GET请求的QPS可达10万，而set、list、zset和hash操作的QPS在6K左右。value大小对平均响应时间（avg rt）的影响主要体现在2000字节以上，此时受主机带宽限制较为显著。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

基础：单机 4G内存 1000M带宽 redis版本:5.0.7

测试：单机redis命令avg rt

维度： 100000request，200客户端并发，value字节数分别为20000|2000|200|20

结论： 1. string类型get请求单机qps 10W，set|list|zset|hash 6K上下

2. string类型的value值大小会影响到avg rt，2000字节以下影响很小， 2000以上影响较明显（这个影响其实并非redis本身的性能影响,而是受到主机带宽的影响）

测试数据如下

100000 request + 200 client + 20000 bytes

redis-benchmark -p 6379 -c 200 -d 20000 -t set,get
====== SET ======
  100000 requests completed in 1.50 seconds
  200 parallel clients
  20000 bytes payload
  keep alive: 1

0.00% <= 1 milliseconds
78.57% <= 2 milliseconds
98.17% <= 3 milliseconds
99.94% <= 4 milliseconds
100.00% <= 4 milliseconds
66844.91 requests per second

====== GET ======
  100000 requests completed in 1.84 seconds
  200 parallel clients
  20000 bytes payload
  keep alive: 1

0.01% <= 1 milliseconds
74.28% <= 2 milliseconds
98.94% <= 3 milliseconds
99.67% <= 4 milliseconds
99.97% <= 5 milliseconds
100.00% <= 6 milliseconds
54259.36 requests per second

100000 request + 200 client + 2000 bytes

====== PING_INLINE ======

  100000 requests completed in 0.82 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



82.71% <= 1 milliseconds

99.91% <= 2 milliseconds

100.00% <= 2 milliseconds

122100.12 requests per second



====== PING_BULK ======

  100000 requests completed in 0.70 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



99.00% <= 1 milliseconds

100.00% <= 1 milliseconds

142045.45 requests per second



====== SET ======

  100000 requests completed in 0.73 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



99.57% <= 1 milliseconds

99.80% <= 2 milliseconds

99.80% <= 4 milliseconds

99.81% <= 5 milliseconds

99.89% <= 6 milliseconds

99.97% <= 7 milliseconds

100.00% <= 7 milliseconds

136798.91 requests per second



====== GET ======

  100000 requests completed in 0.78 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



99.51% <= 1 milliseconds

100.00% <= 1 milliseconds

128534.70 requests per second



====== INCR ======

  100000 requests completed in 0.73 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



99.35% <= 1 milliseconds

100.00% <= 1 milliseconds

137741.05 requests per second



====== LPUSH ======

  100000 requests completed in 0.90 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



67.29% <= 1 milliseconds

99.89% <= 2 milliseconds

100.00% <= 2 milliseconds

110497.24 requests per second



====== RPUSH ======

  100000 requests completed in 0.91 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



83.36% <= 1 milliseconds

99.91% <= 2 milliseconds

100.00% <= 2 milliseconds

110375.27 requests per second



====== LPOP ======

  100000 requests completed in 0.88 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



83.14% <= 1 milliseconds

99.98% <= 2 milliseconds

100.00% <= 2 milliseconds

113250.28 requests per second



====== RPOP ======

  100000 requests completed in 0.92 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



89.02% <= 1 milliseconds

99.98% <= 2 milliseconds

100.00% <= 2 milliseconds

108932.46 requests per second



====== SADD ======

  100000 requests completed in 0.84 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



99.36% <= 1 milliseconds

100.00% <= 1 milliseconds

119047.62 requests per second



====== SPOP ======

  100000 requests completed in 0.91 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



88.26% <= 1 milliseconds

99.80% <= 2 milliseconds

99.80% <= 5 milliseconds

99.81% <= 6 milliseconds

99.88% <= 7 milliseconds

99.96% <= 8 milliseconds

100.00% <= 8 milliseconds

110253.59 requests per second



====== LPUSH (needed to benchmark LRANGE) ======

  100000 requests completed in 0.90 seconds


  200 parallel clients

  2000 bytes payload

  keep alive: 1



79.02% <= 1 milliseconds

99.98% <= 2 milliseconds

100.00% <= 2 milliseconds

110741.97 requests per second



====== LRANGE_100 (first 100 elements) ======

  100000 requests completed in 18.81 seconds

  200 parallel clients

  2000 bytes payload

  keep alive: 1



0.00% <= 1 milliseconds

0.00% <= 2 milliseconds

0.15% <= 3 milliseconds

0.16% <= 4 milliseconds

2.97% <= 5 milliseconds

3.50% <= 6 milliseconds

3.93% <= 7 milliseconds

4.16% <= 8 milliseconds

4.32% <= 9 milliseconds

4.43% <= 10 milliseconds

4.85% <= 11 milliseconds

7.49% <= 12 milliseconds

16.05% <= 13 milliseconds

26.19% <= 14 milliseconds

34.11% <= 15 milliseconds

40.97% <= 16 milliseconds

49.35% <= 17 milliseconds

58.60% <= 18 milliseconds

67.79% <= 19 milliseconds

76.37% <= 20 milliseconds

85.12% <= 21 milliseconds

92.67% <= 22 milliseconds

96.11% <= 23 milliseconds

97.71% <= 24 milliseconds

99.25% <= 25 milliseconds

99.70% <= 26 milliseconds

99.87% <= 27 milliseconds

99.93% <= 28 milliseconds

99.97% <= 29 milliseconds

100.00% <

redis性能指标数据