I ran across this really awesome tool yesterday while reading up about bluehost and upgrading from a large shared server to their “pro plan”.
via: http://www.codemyownroad.com/upgrading-bluehost-pro/

Awesome tool to find out how many sites are hosted under an IP / shared server host
http://www.yougetsignal.com/tools/web-sites-on-web-server/

The reason I set out to read about this is because I just purchased bluehost’s shared service plan, and I’ve seen a few server errors like 500’s as well as some really bad chokes where it takes up to 6 seconds for a simple resource to be obtained. Those issues are on the extreme spectrum, but they are frequent enough, and I’m sure there’s even more room for worse issues to arise.  What is really bothering me is that the request time on the initial site is always 500-1000 ms, other sites always end up being like 150 ms which is my average ping which is reasonable.

The problem is that the requests times are large yet the site looks like it loads “instantly” because after the first resource is requested all the prior resources load without the horrible request times, so once the page starts going it looks instantaneous, yet the initial load people may see as their fault, and blame it on their computer’s speed, or their connection speed, yet it is bluehost’s fault.

I had a talk with their support to see if they would try and get me to upgrade to their pro plan, or what they had to say about the bad request times.  The support told me it loaded instantly for them(would be funny to find out they were on the local network of the host), and they linked me to a site performance analysis of my site which it ranked a solid B in.  I tried to explain the initial request time issue, but it did not seem to matter, I was pointed to use the cloudflare system, but I do not want to have to hassle with more than I should, the site (especially the html page that you are requesting) should load properly without it all having to be hosted through a CDN, at that point what’s the point of the hosting you’re paying for if it doesn’t even do what it’s supposed to do?

My plan of action is to get my full money back, pay for the domain which they will force me to do anyways, and deal with the hassle of finding another host later after my finals are done.  In the mean time I will be hosting my site through my friend’s hosting at media temple.

This does bring to question,  if cheap plans are simply what you pay for, or if there is a cheap host out there that performs properly.  It seems that you may have to spend upwards to $20.00 a month for proper hosting whether you choose bluehost’s pro plan, media temple’s grid service, or a small VPS level 1 server from hostgator.

Even though we have this sweet tool that can show us how many domains are hosted under an IP there are several problems that you run into which prevent you from properly ranking how well your site will run under that server.

•  Dedicated IP’s
  • If you or other people have a dedicated IP on a shared server you will get an inaccurate count of the actual amount of shared domains running on the service provider’s hardware.
  • Also a single machine may be running a range of IP’s which then means that a shared server is actually hosting way more domains than what would be reported using that online tool.
• Hardware Specifications
  • Just because you compare two hosts and see that both have 1000 shared domains doesn’t mean that they will run the same, one server may have twice the RAM and CPU.
• Server Load
  • You may simply end up on a shared server that has a few domains that hog resources, although most hosts try to combat and handle this issue.
• Greed and Money
  • Cheap hosting seems to be what you pay for, hosts want to fit as many customers onto a server as they can possibly fit while keeping a thin balance between profit and customer satisfaction.