Disk Controller Performance

With the variety storage controllers available on these two boards, we needed a means of comparing performance on the wide selection of controllers. The logical choice was Anand's storage benchmark first described in Q2 2004 Desktop Hard Drive Comparison: WD Raptor vs. the World. To refresh your memory, the iPeak test was designed to measure "pure" hard disk performance, and in this case, we kept the hard drive as consistent as possible while varying the hard drive controller. The idea is to measure the performance of a hard drive controller with a consistent hard drive.

We played back Anand's raw files that recorded I/O operations when running a real world benchmark - the entire Winstone 2004 suite. Intel's iPEAK utility was then used to play back the trace file of all IO operations that took place during a single run of Business Winstone 2004 and MCC Winstone 2004. To try to isolate performance difference to the controllers that we were testing, we used the Maxtor MaXLine III 7L300S0 300GB 7200 RPM SATA drive in all tests. The drive was formatted before each test run and a composite average of 5 tests on each controller interface was tabulated in order to ensure consistency in the benchmark.

iPeak gives a mean service time in milliseconds; in other words, the average time that each drive took to fulfill each IO operation. In order to make the data more understandable, we report the scores as an average number of IO operations per second so that higher scores translate into better performance. This number is meaningless as far as hard disk performance is concerned as it is just the number of IO operations completed in a second. However, the scores are useful for comparing "pure" performance of the storage controllers in this case.

iPeak Business Winstone Hard Disk I/O

iPeak Multimedia Content Creation Hard Disk I/O

It is interesting that the performance patterns hold steady across both Multimedia Content IO and Business IO, with the on-board NVIDIA nForce4 SATA 2 providing the fastest IO, followed closely by the Intel ICH7R and Silicon Image 3132 SATA 2 controllers. It is interesting to note that although the MSI and Gigabyte boards both use the same Silicon Image 3132 controller with the same BIOS and driver set, the Gigabyte trails the MSI configuration in each test.

Overclocking Performance Firewire and USB Performance
Comments Locked

19 Comments

View All Comments

  • smn198 - Friday, September 23, 2005 - link

    Welcome Gary. Look forward to seeing more from you.
  • Gary Key - Friday, September 23, 2005 - link

    smn198,

    Thank you. I really enjoyed doing this article, working with Wes, and having the opportunity to share my experiences with the great members and visitors here at AnandTech. I certainly hope you will be seeing more from me. ;-)

    Sincerely,
    Gary Key
  • Evan Lieb - Friday, September 23, 2005 - link

    Welcome Gary, and have fun!
  • Ecmaster76 - Friday, September 23, 2005 - link

    Looks like it went up early.

    Good read though. At first I was like "Holy $#!+" when I saw the gaming benchmarks, but then they mentioned about the Gigabyte BIOS being effed up.
  • cryptonomicon - Thursday, September 22, 2005 - link

    DRAM Voltage Auto, 1.80V to 2.3V in 0.1V increments


    uh.. but isnt ram like 2.5-2.8v?
  • Pete84 - Friday, September 23, 2005 - link

    DDR2 runs at much lower vdimm than DDR.
  • cryptonomicon - Friday, September 23, 2005 - link

    ah yes of course..

    so much for active cooling then
  • BlvdKing - Thursday, September 22, 2005 - link

    I can't believe the Nforce 4 for AMD supports dual core but the Intel edition only has limited support and no support for the 820.
  • coomar - Wednesday, September 21, 2005 - link

    no a diamond can't beat a royal flush

Log in

Don't have an account? Sign up now