Re: [ANNOUNCE] Apache Hello World Benchmarks - Apache C API, HelloDB

Perrin Harkins wrote:
>
> To answer the original question, I don't think Apache::DBI is much 
> overhead at all.  It amounts to little more than a hash lookup. 
> Certainly less work than the the thread synchronization required for 
> connection pooling.

My only problem with Apache::DBI for a benchmark is its
default ping of the db per connect().  For a benchmark that is
little more that a ping itself, this might not be fair to cause
2 pings.  In this light, I might want to just use a global like:

  $dbh ||= DBI->connect();

I just did a quick test, and these were the results:

Test Name                           Test File Hits/sec  # of Hits Time(sec) 
secs/Hit  Bytes/Hit Mem(KB)
---------                           --------- --------- --------- --------- 
--------- --------- ---------
mod_perl HelloDB MySQL Apache::DBI  my_hellod  451.3      9042     20.04    
0.002216  191       16092
mod_perl HelloDB MySQL Global $dbh  my_hellod  570.2     11418     20.02    
0.001754  191       15296

My question is is this a fair optimization for a benchmark
like this.  I tend to not over optimize the benchmarks as
a rule, but I spent hours ( & 12 lines of code :) )
writing an overly simple database connection pooling manager
for the JSP benchmarks where a standard one seems to be
lacking in JDBC & JNDI, so I'm not sure using a mod_perl
global instead of Apache::DBI would be skewing things.
The PHP test I had to do nothing, connection persistence is
enabled by default.

Thanks for the feedback & still looking for more!

--Josh
________________________________________________________________
Josh Chamas, Founder                   phone:714-625-4051
Chamas Enterprises Inc.                http://www.chamas.com
NodeWorks Link Checking                http://www.nodeworks.com