How reliable is the data that "Nginx is faster than Apache2"? Have any real tests been done?

Question

Nginx

Actually, subject.
The fact is that I somehow raised LEMP, so my PunBB worked noticeably slower on it than on a regular LAMP.
I would like to understand what the joke is :)
Are there any "official" tests on this topic?

Answer 1

well, here a person ran tests, it turned out that for the dynamics of apache he turned out a little faster than nginx, and for static nginx about 2.5 times faster than apache. in general, it depends more on how you cook. if you don’t want to bother, then you can put apache purely for php, and nginx for static and as a reverse proxy for apache. in addition, apache is not only a valuable fur but also .htaccess, which is widely used by popular php engines, so I would even recommend such a scheme.

Answer 2

Are there any "official" tests on this topic?

There are so many different types of configurations/loads that you can find tests with any results.
On the one hand, I advise you to use what you are used to / know, on the other hand, Nginx has been so popular in recent years that you need to know it.
As for "noticeably slower" - I'm almost sure that either there was a difference in the configuration of the computer \ cache \ database, or it seemed.

Answer 3

If you disable the script handler in Apache via .htaccess on the folder with statics, then in general, they will become equal in speed.

Answer 4

Compared to the slowdowns that PHP can have with sub-optimal settings, the difference in web server speed is negligible. It slowed you down, apparently, precisely because of this, and not because of nginx.

Answer 5

At the expense of proofs, I defeated the stereotypical understanding that Nginx is faster than Apache.
In the combination of Apache(mpm event)+PHP-FPM and PCRE2 in my Apache for the rewrite module (yes, my Apache already has PCRE2, although it is not yet in the official versions), this bundle showed a good result better than Nginx.