Comment by anschwa

> Intuitively, I'd expect that NGINX serving static content would tend to scale fairly linearly up to the saturation point, and then get rapidly bad

That was essentially what I saw. I was serving very minimal dynamic content that had some random sleeps. I even reached out to the NGINX mailing list and got a similar response about how they wouldn't expect much measurable difference. However, I still think it would be possible to construct a better test.