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 --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|