Is the result of the AB test concurrency problem poor?

I used ci to build an environment and write code to test the mysql + memcache interface. I only use json data to return about 3-4 K of each request (not much) the test results are wonderful. 400 requests. is the result very poor? Requestspersecond: 401.43 [# sec] (mean) if I directly access... I used ci to build an environment, write code, and test it.
Mysql + memcache
The interface only returns json data. each request may look 3-4 K (not much)

Test results are wonderful 400 requests
Is the result poor?
Requests per secondd: 401.43 [#/sec] (mean)

If I directly access PHP echo, the rps is about 2000 characters.
Once you add the ci framework code to echo a string, rps will become more than 500, and the difference between 400 is not big.
I feel that PHP is better than CI concurrency testing.

Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem?

Does the CI framework need to modify the configuration? how can this problem be solved?

Reply content:

I used ci to build an environment, write code, and test it.
Mysql + memcache
The interface only returns json data. each request may look 3-4 K (not much)

Test results are wonderful 400 requests
Is the result poor?
Requests per secondd: 401.43 [#/sec] (mean)

If I directly access PHP echo, the rps is about 2000 characters.
Once you add the ci framework code to echo a string, rps will become more than 500, and the difference between 400 is not big.
I feel that PHP is better than CI concurrency testing.

Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem? Where is the problem?

Does the CI framework need to modify the configuration? how can this problem be solved?

I think the framework itself is much more complicated than the pure PHP Framework. it is complicated to introduce various types of frameworks. so it is certain to consume more resources and time. · The framework is first designed to improve development efficiency. · so we cannot put the execution efficiency first. · of course, there are some methods to optimize it. · For example, Xcache APC and so on. · then optimize the server ···

You can test baidu, sina, qq, and other Chinese sites with AB to check their RPS.