VML wrote:
> Hi All,
> Our company planning to move from Perl/CGI to mod_perl.
> But when we test, mod_perl is slower than mod_cgi.
>
> I test with these script:
>
> # TESTING SCRIPT 1
> use Time::HiRes qw(usleep ualarm gettimeofday tv_interval);
> $t0 = [gettimeofday];
> # CODE HERE
> $t1 = [gettimeofday];
> $t0_t1 = tv_interval $t0, $t1;
> $elapsed = tv_interval ($t0, [gettimeofday]);
> $elapsed = tv_interval ($t0);
> print "$elapsed sec\r\n";
> And
> # TESTING SCRIPT 2
> use Benchmark;
> timethis (1000,
> sub {
> # CODE HERE
> }
> );
>
> Both result are same, mod_perl is slower than mod_cgi.
Oh, boy, this is a wrong test :) You benchmark the code execution and
not mod_perl against mod_cgi. You should use ab(8) or similar app for this.
To learn how to properly do this and much more see:
http://perl.apache.org/guide/performance.html
p.s. also remember that doing wallclock benchmarking as in SCRIPT 1 on
multitask system can give you very incorrect results, if you load is
different during two subsequent benchmarks.
__________________________________________________________________
Stas Bekman JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide ---> http://perl.apache.org
mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org http://ticketmaster.com
