Out of any type of server that an organization will purchase today, web servers are amongst the most popular. "Web server" is such a generalized term in today's IT environment. Their primary function can be one of many different business requirements: an E-commerce site to sell a company's product, an enterprise web services application performing order fulfillment, corporate intranet, collaboration applications, and the list goes on. Today, web servers are more like distributed application servers than a machine serving static HTML pages from years past. In this installment of the Xeon vs. Opteron series, we're going to take a look at both of the most current offerings from Intel and AMD on three different web platforms.

The one common link between the different types of web applications is the language in which they are written. In the past, we've performed our load testing on one application server (Macromedia ColdFusion); however, in this series of tests, we've included three different platforms. We ran load tests on Macromedia ColdFusion MX 6.1, PHP 4.3.9, and Microsoft .NET 1.1. We used a collaboration application called FuseTalk for our .NET and ColdFusion tests, and we used the popular open-source portal software, PHPNuke for our PHP test. We're hoping to get a real world store front in future tests, to further diversify our testing.

Testing procedure

To load up our servers and applications, we used Microsoft ACT, which is included with Microsoft Visual Studio.NET. A test scenario was created that kept the CPU sustained at 90%+ usage, while not completely flooding the box to the point where everything was queued. The test was run over a Gigabit network to ensure that there were no network bottlenecks, and a separate database server was used for all tests. We used Microsoft Windows 2003 Server Web Edition for the operating system, and therefore, had 2GB of memory for all web servers. 2GB or less is a common configuration for a web server today. We used an IDE drive for the web servers, since none of our tests are I/O intensive.

The test results include 3 measurements: Average time to last byte, Total requests served, and Requests per second. The average time to last byte is the average time that it took to receive the last byte of information from each request. The total requests served is the number of successful requests (HTTP status 200) completed within the test time. The requests per second measurement is the average of requests per second that the test was performing throughout the duration of the test.

Opteron System
Dual 250 Opteron processors
2GB PC3200 DDR (Kingston KRX3200AK2) memory
Tyan K8W motherboard
Windows 2003 Server Web Edition (32 Bit)
1 x 40 GB 7200RPM 8MB Cache IDE Drive

Xeon System
Dual 3.6GHz Xeon processors
2GB DDR2 memory
Intel SE7520AF2 motherboard
Windows 2003 Server Web Edition (32 Bit)
1 x 40 GB 7200RPM 8MB Cache IDE Drive

ColdFusion Test Results
Comments Locked


View All Comments

  • Saist - Monday, October 18, 2004 - link

    I just wanted to comment that as I read the graphs, the opteron lead all but .NET
  • justly - Monday, October 18, 2004 - link

    22 There was a thread here that talked about the infoword article, I suggest you at least lokk at it before you claim there are no holes in its conclusion.


    Like any other thread it does drift of topic at times but there are a lot of conclusions made by the author of the article (rck01) that he either could not or would not elaborate on. Personally I wouldnt trust the results of that article untill some of the questions asked in that thread get answered.
  • Chuckles - Monday, October 18, 2004 - link

    No, lets hope that AMD responds by getting a compiler team and optimizing the binary daylights out of the various compilers in addition to boosting the clock. That way everybody wins.
  • dollar - Monday, October 18, 2004 - link

    - The Xeons and the Opterons have the same 1MB L2 cache, cached up Xeons? Huh?
    - There are no Windows Server 2003 64-bit edition but the betas, and people looking for servers most likely aren't using betas. Also there is little or no real software availible. Plus, the Xeons are also 64-bit capable (so that would't put the Opterons at any real advantage anyways.
    - The fact that software is optimized for Xeons turns the tide in their favour, but it's nothing wrong with that, the costumer still gets more performance. Unfortunatly for AMD there isn't likely to be any widespread Opteron optimization with a 5% markedshare. That's what it's like to be the little guy :(

    Xeon's came out victorious (for once), let's hope AMD responds with faster CPUs ;)
  • sprockkets - Monday, October 18, 2004 - link

    I guess my only question is what type of Xeon was used, the ones with huge cache, or what?

    I liked that Infoworld report, in a way. It's weird that Opterons are faster, but load up a Xeon with work and it is faster. Kinda makes sense, but if you are faster when under load, wouldn't you run faster without load? Does that mean Xeons waste most of the time without other programs to take advantage then of Hyper Threading?

    Then again, with a shared bus architecture it's been proven here until the Xenon went up to a 800mhz bus that Opterons scale much better.
  • Guspaz - Monday, October 18, 2004 - link

    Isn't it kind of an unfair test to compare a 64-bit processor in 32 bit mode to a 32 bit processor?

    If you're going to benchmark without taking advantage of one of the Opteron's biggest features (AMD64), why not disable something like Hyper-Threading to compensate? From what I understand both AMD64 and HyperThreading provide similar performance increases of something like 5 to 15 percent.
  • WooDaddy - Monday, October 18, 2004 - link

    I understand for the purpose of the review, it might be easier to order from the same vendor, but since the architecture of each one is different (AMD/Intel), most companies wouldn't sell conflicting yet equivalent servers. The server market caters less to fanboys/men/women than the fickle consumer market.

    My point is that if you are going to compare full blown pre-assembled systems, you should look at competing vendors since it isn't cost effective for the vendor to release self-competing products. But since you probably built your own (the test configuration info was rather sparce), make sure the component cost is the same. That's the comparision that would count.
  • Jason Clark - Monday, October 18, 2004 - link

    I plan on adding a cost comparison in our next article in this series. The trick is to make it as fair as possible where each system is outfitted in a similar fashion from the same vendor if possible.

  • Questar - Monday, October 18, 2004 - link

    Oh yeah, that's a really good idea. I should base business buying decisions based upon a review at a gaming site. There's a source I would quote in my research.
  • mrdudesir - Monday, October 18, 2004 - link

    That review at info world was a joke. No Data, just assertions that they claim, would they care to back it up with evidence. If you want to see a real review check out this one over at GamePC that i found, which actually outlines all of the hardware and apps used.

Log in

Don't have an account? Sign up now