> In Apache, each request is handled by 1 OS thread, effectively capping Apache to thousands of concurrent connections That's wrong. There are a variety of di

> In Apache, each request is handled by 1 OS thread, effectively capping Apache to thousands of concurrent connections That's wrong. There are a variety of different Apache mpm's: https://httpd.apache.org/docs/2.4/mpm.html https://httpd.apache.org/docs/2.4/mpm.html In the footnotes, it also points out that Erlang uses a similar system, which is true, and worth a look as well.