The supposed issue with cgi performance is fork(), but for is much faster than people think, at least on Plan 9 and (specially for statically linked binaries) on Linux.
As Ethan found out, werc performance is mostly limited by whatever fork() sucks in your OS, and I/O. Werc seems to beat the shit of most supposedly faster systems like php or python with fcgi or scgi. Note that perl(or python) startup time alone is much slower than what takes werc to serve a request (which includes many forks), by far the slowest component of werc is the default markdown.pl, which is why we cache its output or why you should use a different implementation of markdown which is in a language that doesn't suck. Put another way: one of the purposes of werc was to prove that the myth that cgi and fork are slow is bullshit, and I'm very proud that werc proved this well beyond my expectations. Get an OS where fork() is not totally broken and make sure your binaries are statically linked and you are not using retarded bloated languages like perl, or python (or god forbid, Ruby: http://harmful.cat-v.org/software/ruby/ ), and you will be fine. Uriel On Tue, Jul 3, 2012 at 1:41 AM, quiet_lad <gavcom...@gmail.com> wrote: > curious > > -- > You received this message because you are subscribed to the Google Groups > "werc" group. > To post to this group, send email to werc9@googlegroups.com. > To unsubscribe from this group, send email to > werc9+unsubscr...@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/werc9?hl=en. > -- You received this message because you are subscribed to the Google Groups "werc" group. To post to this group, send email to werc9@googlegroups.com. To unsubscribe from this group, send email to werc9+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/werc9?hl=en.
