怎么都测不过，一次请求要5、6秒，我着实惊讶
1 经过多翻调查，先ping redis和drds，发现ping值要3毫秒左右，经过调整vpc到一个网段后，解决0.x毫秒为正常ping
2 将for使用parellel实现并行
3 自测发现2百毫秒，而在压测环境在6百毫秒，经过调查发现压测环境用的debug级别日志，通过阻塞io增加了延迟，这是一个大坑
自测命令yum install httpd-tools，然后通过
ab -n 1000 -c 80 -H 'resourceScale: 1' 'http://xx/api'
Requests per second: 566.39 [#/sec] (mean)
Time per request: 176.556 [ms] (mean)
Time per request: 1.766 [ms] (mean, across all concurrent requests)
Transfer rate: 2385.05 [Kbytes/sec] received
Connection Times (ms)
min mean[+/-sd] median max
Connect: 0 0 0.3 0 17
Processing: 48 175 22.2 173 342
Waiting: 45 175 22.2 173 342
Total: 48 175 22.2 173 344
Percentage of the requests served within a certain time (ms)
50% 173
66% 180
75% 186
80% 190
90% 204
95% 217
98% 232
99% 244
100% 344 (longest request)
其中第一个Time per request为平均用户感知速度，应该是可用的 90%是压测team值用的参考值。
done