Hi Angel,
No problem, its a good question and something that we will eventually
cover in our docs (which are admittedly very sparse at this point).
For the mean time since the underlying web server is tornado you can
refer to their production deployment docs here:
http://www.tornadoweb.org/documentation#running-tornado-in-production.
We run our whirlwind apps with the same setup they suggest and it
works very well. That said you could potentially use apache with
something like mod_proxy_balancer as well if thats more to your
liking. We like nginx because its fast and lightweight.
One thing that gets a bit annoying with running tornado apps is
starting up all the instances that your load balancer to proxy too.
There are bunch of approaches that people have come up with. Most of
which are overcomplicated. So we created an init.d script that should
help with this. Look here
https://github.com/trendrr/whirlwind/tree/master/etc.
Basically you have a config file that lives in etc. This defines the
ports you want to start whirlwind instances on and then a start up
script that lives in your /etc/init.d directory. Then its a simple
matter of:
/etc/init.d/whirlwind start
/etc/init.d/whirlwind restart
/etc/init.d/whirlwind stop
Please note that this will not work on windows.
As for not needing apache (or some sort of other load balancer) you
could in theory start whirlwind on port 80 but to be honest I wouldn't
suggest it. The tornado docs cover a few reasons why you wouldn't want
to do this.
Hope that helps and thx for checking out whirlwind. :)
best,
Matt