"Pancake is a fast, simple and lightweight HTTP-Server written in PHP with acceleration for PHP-powered websites"<p>So...it's written in PHP, so it loads up a PHP application and serves it faster than running through a real webserver to a PHP backend via FCGI. Duh?<p>Now go ahead and add in high concurrency and static asset serving and add PHP-APC to the workers and don't benchmark on an OpenVZ VM (which doesn't guarantee resources!) and let's see how things shake out.<p>Pancake seems to be touting this "CodeCache", which is great, except that PHP already has APC, which does this already, and I'll wager it does it better. This smells like a false benchmark, where Pancake wins because it's avoiding work that the poorly-configured nginx stack has to do. If you were to tune both of them to their maximum potential, I would be shocked if nginx didn't eat pancake for breakfast.
The license for this web server is quite unusual,<p><a href="http://pancakehttp.net/license/" rel="nofollow">http://pancakehttp.net/license/</a><p>I would expect a more sensible license.<p>P.S, that license page is so well hidden.. only possible to find by going to the github page
Interesting but: This license does not allow using Pancake for commercial websites. If you want to use Pancake for such sites, please contact support [at] pancakehttp [dot] net
Can we really trust numbers returned from a benchmark run on a VM? It seems like the other load on the system during each benchmark would effect the results.