the-benchmarker / web-frameworks

Which is the fastest web framework?
MIT License
6.91k stars 641 forks source link

Update config for simps #3104

Closed sy-records closed 3 years ago

sy-records commented 3 years ago

Hi, @waghanza Can you test it for me?

waghanza commented 3 years ago

Nice improvement

Could you please make the same for all swoole based frameworks ?

Language Framework Speed (64) Speed (256) Speed (512)
Before php (7.4) simps (1.0) 120 776 155 576 177 734
After php (7.4) simps (1.0) 108 843 170 021 188 892
sy-records commented 3 years ago

Not necessarily applicable to all Swoole frameworks 🤣

waghanza commented 3 years ago

are not swoole options ?

sy-records commented 3 years ago

It is a configuration of swoole, but not necessarily useful for other frameworks, I'm not sure.

Merge it later and I'll try to test it again.

waghanza commented 3 years ago

Ok. If you have no time to do so, I think I will try later :stuck_out_tongue:

My purpose is to keep (try to) same behaviours for different swoole based framework, I mean having as-far-as possible comparison

sy-records commented 3 years ago

What is the linux kernel of the test environment? @waghanza

sy-records commented 3 years ago

Can you run again?

waghanza commented 3 years ago

What is the linux kernel of the test environment?

5.7, I use the same distro than for cloud => fedora

Can you run again?

No. I'm not on a computer. But I can help you :stuck_out_tongue_closed_eyes:

sy-records commented 3 years ago

Well, you can run it when you have time.

sy-records commented 3 years ago

Hi, @waghanza Do you have time now? Can you run again?

waghanza commented 3 years ago

Since my daughter is born I have time @sy-records :stuck_out_tongue:

I find :

Language Framework Speed (64) Speed (256) Speed (512)
Before (master) php (7.4) simps (1.0) 104 286 144 225 165 106
After (this branch) php (7.4) simps (1.0) 108 460 169 109 188 488
sy-records commented 3 years ago

Congratulate you on becoming a father. 😆

These configuration items seem to have little effect, and the results of each test are different...

waghanza commented 3 years ago

Congratulate you on becoming a father. laughing

I'm already one, but I've lost my skills 😉 btw, thanks

These configuration items seem to have little effect, and the results of each test are different.

Due to docker ... There is too much fluctuations to consider something I suggest to wait for the cloud version to test this 😚 or if you have an other idea ....

sy-records commented 3 years ago

When can you use the cloud version for testing?

The results is using a cloud version, right? You can also merge first, I can wait for the results to be updated. :)

waghanza commented 3 years ago

When can you use the cloud version for testing?

I don't know, it depends on my own schedule, and as you may know, this is not my top priority, right now

Their is no cloud results published for now, take a look at the roadmap https://github.com/the-benchmarker/web-frameworks/issues/3108

sy-records commented 3 years ago

https://github.com/the-benchmarker/web-frameworks/pull/3068#issuecomment-652254170 In this comment, I thought you had used the cloud...

Is it still running in a docker environment?

waghanza commented 3 years ago

Yes. Cloud is not yet ready ...

laravel has too many files, dart is failing ... are some issues on cloud but I think I'll publish a cloud version in September

I'm still running in docker as for now. I can try to generate docker based results, along podman, but I think it will have the same issue

sy-records commented 3 years ago

Okay, let's merge.