MySQL Lists are EOL. Please join:

List:General Discussion« Previous MessageNext Message »
From:Tom Cunningham Date:June 23 1999 5:51pm
Subject:Re: Persistent connections to db?
View as plain text  

On Wed, 23 Jun 1999, Paul DuBois wrote:

> At 12:48 PM -0400 6/23/1999, Tom Cunningham wrote:
> >I'm running Red Hat 6.0, and I'm unsure of whether I'm using persistent
> >connections to my mysql db. I've installed the MySQL binaries on my
> >system. How would I find out whether I'm using persistent connections or
> >not, and if I'm not, how would I go about setting up persistent
> >connections?
> 
> Persistent connections are a PHP thing.  Are you using PHP?


Nope. Perl and mod_perl. Someone suggested my problems might be persistent
connections, so I wanted to explore that problem, but I'm not so sure
about that. I'm using bind_param and bind_col everywhere I can in my CGI
scripts.

I'm configuring a new Red Hat 6.0 machine with SCSI drives that I've
determined through benchmarks to be 2 times as fast as my current Red Hat
5.2 server. I've run the mysql benchmarks on both machines, and the new
machines has come out roughly twice as fast. I've ran Apache benchmarks on
the new and old machines using static html, and the new machine seems to
be twice as fast at that. Then new machine also is about twice as fast at
doing non-db driven cgi-bin perl, and non-db driven mod_perl.

When I try to benchmark my cgi though (exact same script on both machines,
I get the strange results :

From Apache Benchmark :
                        Old Machine             New Machine
Requests per second :   1.47                    .84
Processing Min          1980                    1940
Processing Avg          2665                    4685
Processing Max          4273                    6869

I ftp'ed the tables straight from the old machine - which is RedHat 5.2
and the new one is RedHat 6.0. Would glibc factor into this at all? Would
I be better off creating the tables locally on the new machine?

Every benchmark I've ran says that the new machine is faster (which it
is), except the CGI I really need to be faster. The wide variance in
processing time in the Apache Benchmarks for the cgi on the new machine
really make me wonder what's going on, and whether there's some library
issue here.


Thread
Persistent connections to db?Tom Cunningham23 Jun
  • Re: Persistent connections to db?Paul DuBois23 Jun
    • Re: Persistent connections to db?Tom Cunningham23 Jun
      • Re: Persistent connections to db?Daniel Koch23 Jun
        • DBD::mysql fails on two tests?Tom Cunningham24 Jun
          • glibc 2.1 performanceTom Cunningham25 Jun
            • Re: glibc 2.1 performance(Paul D. Smith)25 Jun
              • Re: glibc 2.1 performanceTom Cunningham25 Jun
                • Re: glibc 2.1 performanceathompso28 Jun
    • Re: Persistent connections to db?ed23 Jun
Re: glibc 2.1 performanceAndrew Dunstan25 Jun