Different Web Hosting Options
VPS
Virtual private servers are a great option if you want the power of a dedicated server without the cost. The downside I have found is that your website is still hosted on the same physical machine as other sites and I have almost certainly written some infinite loop code that has nearly brought a machine to its knees. While that's not supposed to be an issue since every customer runs in their own space on the machine, in practice I have seen it cause issues.
Dedicated
With a dedicated server, you have your own machine and you won't have to be concerned about the websites of other customers at the hosting company degrading the performance of your site. There are 2 flavors of dedicated server, fully managed and root server. The root server is the most powerful because you can modify anything on the machine. The downside is that you can really mess up the machine if you don't know what you're doing and then you'd have to do a reset which might lose some of your work. You also have to take care of upgrades and security on a root server and unless you've got some LINUX and sys admin experience, these tasks can be daunting.
Cloud
Hosting your site in the cloud is the latest and greatest option and one I particularly like for websites that may get a lot of users or that are doing very heavy data processing. Again, there are a couple of options with cloud hosting, one is Google App Server which abstracts most of the complexities. All you need to do is deploy your Python code on their system and they handle all reliability and scalability issues. The downside is that you have to write your site in Python and that you're only supposed to use it for the app's front-end, i.e. for serving webpages loaded by humans. My preferred option is Amazon-style cloud hosting because they provide you with all the building blocks to create any web application. The downside is that you need to put the blocks together and this has a steep learning curve, but once you're comfortable you could start up a web server, load balancer, fully redundant database and some back-end machines for processing data (like indexing) in a matter of hours and not weeks.
