The unicorn talks about this for "fast customers." Whenever I read a detailed description of this, it is explained by the fact that it is suitable, for example, for the nginx proxy server serving HTTP requests, and not for launching an application that does a long poll.
Well, it makes sense, but the more common case when the request is “slow” is when the POST takes a lot of time, either because it loads a lot of data (file download), or because the action causes processing in an application that takes a lot time.
How does a unicorn behave in these two cases? Where can I find out more about this?
source
share