Marc Battyani <Marc.Battyani@fractalconcept.com> wrote:
+---------------
| "Rob Warnock" <rpw3@rpw3.org> wrote
| > Footnotes:
| > [1] Not exactly the same as "mod_lisp" <URL:http://www.cliki.net/mod_lisp>,
| >     but close: a *tiny* C-coded CGI trampoline connects to the CMUCL image
| >     via a Unix-domain socket, then speaks a "mod_lisp"-like protocol to it.
| 
| Tss, tss, a CGI trampoline...
| mod_lisp is the only mod_lisp ;-)
+---------------

Yes, yes, I know. (*blush*)  But I at the time I wrote it there were
acceptance issues which made it easier to use Lisp if the existing
Apache server didn't have to be modified (or even restarted!) to add
the Lisp server. And the CGI trampoline ran plenty fast enough on the
target platform, ~100 requests/sec (as measured with "ab"), while
the user load was never expected to be more than ~100 requests/HOUR!
[For comparison, on the same machine, a trivial Perl CGI script got
only ~140 req/sec, not that much better.]  So it seemed a "good enough"
tradeoff.

And once the Lisp server has been running for a while, well, one can
always go back and say, "You know, we could speed this up a bit if
you'd let me install this simple little Apache module..."  ;-}  ;-}


-Rob

-----
Rob Warnock			<rpw3@rpw3.org>
627 26th Avenue			<URL:http://rpw3.org/>
San Mateo, CA 94403		(650)572-2607