[Catalyst] Catalyst performance under load

[Brandon Black]

On 2/9/06, Richard Jolly <Richard.Jolly at bbc.co.uk> wrote:
>
>
> Hi,
>
> We've been using Catalyst for a project with much happiness. But now we have
> some hostile management questions about performance under load. Basically
> its considered new, unproven technology good for 'prototyping' but quite
> probably inappropriate for production.
>
> So I'm asking if there is any existing data for load/stress testing, or
> anecdotal accounts of it being used for high availability/high load sites.
> I've seen the wiki page listing sites using catalyst [
> http://dev.catalyst.perl.org/ ], but I fear none of these are big enough to
> convince management.
>
> We have to make a case to an unsympathetic audience - please help us gather
> evidence and arguments.
>

Like everyone else said, Cat is a lightweight framework, it's unlikely
to be the source of any serious performance issue.  If you hit any
performance limits in testing, it will probably be in your ORM, your
database itself, or your application code.

For ORM issues, if it's just one hugely complex set of queries
somewhere that bogs down, you can always convert just that part to
straight-DBI calls, and let someone on the respective mailing list
know about it so they can try to improve that situation.  If it's the
database itself, well, that's a whole different issue.  If it's your
app code, it's your app code, you fix it.

Aside from all that, if you end up with a really lightweight
super-performance Catalyst app, and you're looking to squeeze every
last drop out of it, I would bet switching from apache+mod_perl to
lighttpd+fastcgi could gain you a few more connections per second.

Luckily for you, Catalyst is such a flexible framework, that making a
server architecture switch like that doesn't involve modding any of
your code.

-- Brandon