Closed jamesli2021 closed 11 months ago
Hi @jamesli2021,
Thanks for the idea. Not sure this matter in fact ...
Let me explain. We are benchmarking web frameworks.
The compiled / java-based languages has built-in servers (go
, kotlin
...).
nginx
is used only for some interpreted languages (ruby
, python
, ...).
The most important for me, is results consistency. If we want to compare for example ruby
and python
(could be legitimate somehow) we should compare only if they are using the same http layer engine, no matter what it is.
I've taken nginx
since it is really a swiss-knife, but OpenLiteSpeed could have been choosen.
For me this idea could be implemented only after runtime feature https://github.com/the-benchmarker/web-frameworks/pull/3958
This will then allow results to be shown either with nginx or with openlitespeed (we also can use apache or litehtttp or passenger ...).
Regards,
LiteSpeed Enterprise is widespread across web hosting that could be interesting when OpenLifeSpeed (non-enterprise) is use in the benchmark comparison.