Page History: Hosting benchmarks
Compare Page Revisions
Page Revision: 2010/01/30 16:12
DoulaMatch.net keeps growing, as does our Facebook app BirthWatch, so I am always looking at whether our hosting needs are keeping up with our business requirements. I've always used shared hosting plans in the past. Very cheap, but you do get what you pay for. One of the issues looming for us was that Facebook was going to make us send our own emails to our app users, which I could see quickly exceeding the typical email usage limits on most shared hosting plans. SoftSys Hosting, our current host, limits us to 100 per hour. After you exceed this, your account is automatically suspended, which would be a very bad thing.
One of the reasons I chose to switch to SoftSys recently was that they also offer reasonably priced Virtual Private Server (VPS) plans. So I fired one up to migrate our big apps to. I am looking forward to the possibility of better performance as well, which is the point of this page.
I'm going to benchmark four scenarios:
SoftSys shared hosting for SQL Server and IIS
SoftSys VPS for IIS and shared hosting for SQL Server
SoftSys VPS for IIS and SQL Server (all on the one VPS)
SoftSys VPS for IIS and our old host, ReliableSite.net for shared hosting SQL Server
My test will be getting 4 different Doula search pages, for 5 iterations each, from DoulaMatch.net. I'm going to use the new DoulaMatch code base which uses MVC. All of our data access is through LINQ.
Here are the test search URLs:
/list/birth/wa/5 (5th page of Washington birth doulas)
/list/birth/ca/4 (4th page of California birth doulas)
I'm going to use the Hammerhead Firefox plugin to perform the tests to get a true end-user benchmark.