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
POST A COMMENT

47 Comments

View All Comments

  • Questar - Monday, October 18, 2004 - link

    #25,
    RTFA, the numbers are all there. Now who's the joke?
    Reply
  • Phiro - Monday, October 18, 2004 - link

    Good article - we currently exclusively use clustered dual-P4 cpu boxes w/hyperthreading enabled & Windows 2003 for our web clusters, and the .net side of it gets larger every day, so this quick review was right on the money. The one thing I would have added was a quick cost comparison for the like servers.

    People, you may all beg and cry for things like Linux testing, Windows 64 testing, etc. etc. but given the simple constraints on this article - basic web server comparison using hardware & OS's and software that's actually in production at most Fortune 500 companies - that's useful to some of us. I'm sorry they didn't benchmark this with Raptors, 42 sticks of ram and the latest beta nvidia drivers, but this wasn't a "how high can we max out web server performance" - they write those on odd-numbered days :)
    Reply
  • PrinceXizor - Monday, October 18, 2004 - link

    That infoworld "review" was a joke. No benchmarking numbers, just the reviewers own subjective opinions. There may be some grain of truth in the multi-tasking vs. non multi-tasking strengths of the Xeons. But, its impossible to say from that "review". There was virtually no useful information for evaluating the platforms. Its no wonder that most useful reviews are from on-line hardware sites like AT, [H], TechReport, THG, etc.

    P-X
    Reply
  • WooDaddy - Monday, October 18, 2004 - link

    BIG problem with this review. HOW MUCH DO THEY COST!?!

    No self-respecting company would purchase a server without performing a cost/performance ratio.

    My assumption is that the Opteron system is less expensive.. that being said, then the review is moot. Equally priced systems should be used. My next assumption is that then the new Opteron system would spank the Intel...

    But unfortunately, I'm an AMD fanman.
    Reply
  • Questar - Monday, October 18, 2004 - link

    That should say that testing of Opteron vs Xeon with workstation class loads has already been done by others Reply
  • Questar - Monday, October 18, 2004 - link

    Testing of Opteron vs Xeon has already been done by others:

    http://www.infoworld.com/article/04/08/13/33TCwork...

    To quote:
    "What we found was eye-opening. The Opteron machine outperformed the Xeons when lightly loaded with minimal multitasking, but once the real work started, the Opteron stopped. It was effectively shut down by the same multitasking load that the two Xeons performed with ease. In the clean environment, it still performed at less than half the speed of the older and allegedly less-capable Xeons."
    Reply
  • allnighter - Monday, October 18, 2004 - link

    I feel sorry for all my fellow AMD fans for getting so defensive over one test result. Especially when clearly stated that the application uses optimized code for Intel processors. What's the big deal? Anyone surprised? Reply
  • Jeff7181 - Monday, October 18, 2004 - link

    #19... since you obviously didn't read the other posts or the article and just looked at the graphs... the fact that the Intel solution was faster than the AMD solution in .NET is more likely due to the fact that .NET has been optimized for the P4. Reply
  • MightyB - Monday, October 18, 2004 - link

    I think it is nice to see Intel finally beat a opteron .. they only had to make a 3.6Ghz and beef up the cache before they could match it :-).. The only thing this review lacks like all others is to mention the prize difference.. how much more do u pay to get those 8% in .NET....??

    Best regards
    LittleB69
    Reply
  • Rohde - Monday, October 18, 2004 - link

    #10 - Since they only used a single pair of memory sticks, Numa would not have made a difference.

    If instead they attached 1GB of memory to each processor, we probably would have seen better performance in all benchmarks since one CPU would not be taking a memory latency hit with each access.
    Reply

Log in

Don't have an account? Sign up now