If you are testing, can you do it again with putting the .htaccess in cache? I am just curious :)
https://hoytech.com/vmtouch/
>
> The HTTPD documentation says "You should avoid using .htaccess files
> completely if you have access to httpd main server config file. Using
> .htaccess files slows down your Apache http server. Any directive that you
> can include in a .htaccess file is better set in a Directory block, as it
> will have the same effect with better performance."
> I wanted to see if I could measure how much slower it is and boy does it
> make a difference. (This is actually part of my PhD research into how to
> better understand configuration management.)
>
> I built HTTPD from source with a lot of debugging features (i.e. symbols, no
> compiler optimization), so these specific numbers are only valid in the
> context of this test, but they are still interesting.
>
> I created a file DOCUMENT_ROOT/1.txt containing just the text "1" (Short
> URL) and a file
> DOCUMENT_ROOT/1/2/3/4/5/6/7/8/9/10/11/12/13/14/15/16/17/18/19/20/21.txt
> containing the text "21" (Long URL).
> I used ab to run get requests, first 10,000, then 100,000, then 100,000
> again, just to check for variability (and it turns out there isn't any
> significant variability). I ran ab on the same machine as httpd.
> I ran the requests first with AllowOverride None, then with AllowOverride
> All but no .htaccess files, then with AllowOverride All and .htaccess files
> with a mix of "Require all denied" and "Require all granted".
> I also collected the number of instructions executed by the system using
> perf.
>
> Here is a quick summary of the results:
>
> AllowOverride None / Long URL:
> - 1.367 seconds
> - 13.637 seconds
> - 13.607\ seconds
>
> AllowOverride None / Short URL:
> - 1.283 seconds
> - 12.981 seconds
> - 12.989 seconds
>
> AllowOverride All / Long URL:
> - 2.002 seconds
> - 20.015 seconds
> - 20.032 seconds
>
> AllowOverride All / Short URL:
> - 1.370 seconds
> - 13.581 seconds
> - 13.590 seconds
>
> AllowOverride All / Long URL with `.htaccess` files:
> - 3.062 seconds
> - 31.042 seconds
> - 31.122 seconds
>
> AllowOverride All / Short URL with `.htaccess` files
> - 1.431 seconds
> - 14.487 seconds
> - 14.461 seconds
>
>
> The change in perf counters matched the changes in wall clock time.
> The only thing surprising about any of these results was the magnitude of
> the performance effects.
> I think it is most interesting that for this example of a path 20
> directories deep, having an extra .htaccess file nested in each directory
> actually doubled the amount of time it took to process the request.
>
> - Y
|