Posts Tagged ‘Web Servers’

October 1st, 2007

Anything For Sale By Owner

Anything For Sale By Owner LogoAs I alluded in a post a couple of weeks ago, I have been a bad blogger. And I have neglected my community of readers. However I would like to tell you what I have been doing in the last couple of months while I have been neglecting my blog.

I recently got involved in creating a startup as the lead developer for an online classifieds site called Anything For Sale By Owner. From the ground up this was conceived as a middle-ground between craigslist and ebay where every listing would be charged at a static rate of $1.00/month. The $1.00 is a way to week out the crap from craigslist and the death-by-fees from ebay.

The description is pretty straight forward but the more interesting tid-bits that I know my readers are interested in are as follows:

.NET/C#

We choose .NET mostly because it is what I knew and .NET for me has always been stable, predictable, and performed really well on server-side applications. The alternative was PHP and even through we wrote many of the processing layers by our self (i.e. REST Web Service Handler) the time to deployment was greatly accelerated because of all the work the Microsoft ASP.NET Team has put in to the product. The user of Master Pages and Web Services made for an easy separation between content and display.

MySQL

We choose MySQL for a whole host of reasons mostly based around the costs associated with a single Microsoft SQL Server Standard Edition license. Other reasons we choose MySQL was for scalability, because not only could we install 5 database servers (hardware included) for the same price we could purchase 1 MSSQL database server for but also because the master/slave replication of the databases seemed to be an easier process when we needed to scale horizontally.

REST Web Services/AJAX

We followed the Digg Model for exposing web services and each web service could be changed around to provide output through JSON, RSS, ATOM, KML (where applicable), and XML. I even did a write up about a month ago on the JSON Serializer that I developed for this website. This was very important for the AJAX we needed to control many aspects of the user experience.

Open Search

Open search was one of the value add features that wasn’t in the original spec but was an easy add-on because we already had the Web Services in place to leverage. You can view our open search XML definition file here. If you are unfamiliar with Open Search this is how A9.com defines it:

OpenSearch is a set of simple formats for the sharing of search results. Any website that has a search feature can make their results available in OpenSearch format. Other tools can then read those search results.

So that little search box in the upper right hand corner of your browser is an example of an Open Search client.

SEO

Search Engine Optimization is a very important part of any website today. Because for most new sites and even some of the older ones, Google is going to be the largest most ignored user of the site. Not only will Google look at every single page on your site every week, which I dare any human to try and accomplish, they will also be the largest organic promoter of your site.

One of the corner stones of SEO is easily readable URL’s that contain descriptive keywords. This means you have to have a good URL Rewriter in place. We choose Ionic’s ISAPI Rewriter that integrated nicely with IIS 6.0. However that left a big gap between using Visual Studio’s Intigrated Web Server (which doesn’t support ISAPI) and a full blown IIS Server. The benifits are pretty obvious for running the Intigrated Web Server that comes with Visual Studio, for one you don’t always have to have IIS that comes with Windows XP always running and the host of security problems that comes with it, two I was running Windows Vista and IIS 7.0 has some huge differences from IIS 6.0.

So I sat down one night and wrote my own Apache mod_rewrite compatible HttpModule for running the same rule-set that I defined for Ionic’s ISAPI Rewriter, to fill in the gap and make developing on my local machine as close as running the live web site on IIS 6.0.

If anybody is interested I offer the Url Rrewriter as a free download:

Front End Design

I am a software developer and usually don’t get invovled in the artsy end of the web site design. So I will let my co-worker Tom Lauck describe how he developed the front-end for Anything For Sale By Owner.

So all in all I believe that this website has a very good chance of making in the Wild Wild West that the Internet is, however that is probably just the ramblings of a proud father. Be on the look out for some major marketing campaigns in the NYC Times Square region.

Tags: , , , , , , , , , , ,

Posted in ASP.NET, C#, Portfolio, Programming, Rant, SEO, SQL | kick it on DotNetKicks.com | Bookmark | View blog reactions | 1 Comment »

June 30th, 2007

Which web server is better under load, IIS 6 or Apache?

One of the many techno-geek religious arguments that comes up a lot is which web server has a faster response time under load, IIS 6 or Apache? I am happy to say somebody actually put this to a test using what is known as the Digg-effect, basically a constant hammering of the server to keep it under load. The results may surprise some of the zealots out there and the test might be buried because of an unpopular fact. Here is the setup from the site:

This is a page to test the effect of high reddit and digg hits on two different servers one running IIS6 and the other Apache. The purpose is to see how each handles high hit loads and is the most reliable.

By using one server to load this page (not being tested) then calling a page from a dedicated IIS6 server into an iframe and a second page from a dedicated Apache server into a second iframe. The entire process is using PHP scripting and mysql data to store the results. To eliminate cache hits on both test servers, the page being returned to the iframe is dynamically created each time from a php script.

After the pages are completely loaded, an ajax call is made to the primary server to record the times back into the sql database for statistics. All three servers are the same physically and in the same rack and network. Bandwidth is not a measurement issue, since only the execution of the php script is being measured.

I have taken the liberty of making a screen shot of the following site just in case it is taken down. The screen show is dated 2007-06-30.IIS 6 vs Apache Graph

I have included the results below for the same reason.

Reddit hits 27653
Digg hits 874
Seconds to call the iframes from main page this run 0.0528259277344
Total seconds to load all pages this run 4.02603888512
Average seconds to load the iframes (both) 2.60272280153
Average seconds to load IIS 2.2937795829352
Average seconds to load APACHE 2.9116660201344

This is a very interesting study, and I am going to keep following this site for updates.

Tags: , , ,

Posted in Programming | kick it on DotNetKicks.com | Bookmark | View blog reactions | No Comments »