Ruediger Pluem wrote:
On 11/02/2005 11:54 PM, Bill Stoddard wrote:
[..cut..]
Yep, you see it correctly. Mod_mem_cache is most interesting to me
personally as a benchmark toy. It should be possible to extend
mod_mem_cache to cache headers (rather than rebuilding them for each
response sent out of the cache), align storage on processor cache-lines
and other performance tweaks to serve static pages really really fast
But in this case you have the situation that you have to copy the memory
contents between user and kernel space. Just asking: Does this beat
a working sendfile implementation on a system with filled OS buffer cashes?
That's where caching open file descriptors comes in handy.
(or cache dynamically generated pages until they become stale). You
should be able to use mod_mem_cache to explore the static page
performance limits of Apache HTTPD 2.0.
Thanks for the clarification. So I would think the effectiveness
of mod_mem_cache grows with a shrinking number of httpd processes.
That said I guess the windows mpm by default should provide one of the
most effective environments for mod_mem_cache in contrast to prefork on
the other end.
Caching file descriptors on windows will work -really- well when we get around to fixing the braindead
implementation of apr_sendfile on windows.
Bill
Regards
Rüdiger
