Re: Perpetuating the myth...annoying

On Mon, 5 Apr 2004, David Costa wrote:

>
> On Mar 31, 2004, at 4:44 AM, Jacob Hanson wrote:
>
> > http://www.oetrends.com/news.php?action=view_record&idnum=319
> >
> > Article discussing PHP5 says SQLite is "typically faster than MySQL,
> > significantly faster than PostgreSQL". While I don't doubt it's
> > faster at what it does (with it's small, focused feature-set), the
> > article perpuates the 'Postgres is slow' and 'MySQL is much faster
> > than Postgres' thinking that is no longer accurate.
> >
>
> Hi Jacob,
> I have been using SQlite in fact we offer it on dotgeek together with
> postgresql in the free
> package for developers. The significantly faster is irrelevant because
> SQLite has a single point of access / failure
> which make it interesting only for small applications.
>
> Just an idea, why don't we run a postgresql vs mysql benchmark ? I can
> put the servers, but I have little experience
> on running a benchmark. If someone is willing to help I will grant him
> access and we can end up with some clean results
> and a nice article on dotgeek with something effective like "PostgreSQL
> is faster then MySQL: Here's why"
>
> I think it will spice it all up. Till now we have around 300 PHP
> developers that started to use PostgreSQL on dotgeek and
> porting their application from mysql to postgresql. Looks promising ;)

I think that if we are going to benchmark the two, and I'm all for it, we
should benchmark them running something fairly standard, like the OSDL
tests. And MySQL should be using innodb tables for everything, since
postgresql pretty much always carries the load of running mvcc and write
ahead logging and transactions for everything it does.