Disk Controller Performance

With the variety of disk drive benchmarks available, we needed a means of comparing the true performance of 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. 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 differences 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.

Our testing results in RAID 5, 0, and 1 with the Intel ICH7R was successful, but we did not have time to properly complete testing of the other storage controllers for this article. We will have published RAID results for all tested boards in our 975X roundup along with results from the NVIDA based P5N32-SLI.

iPeak Business Winstone Hard Disk I/O

iPeak Multimedia Content Creation Hard Disk I/O

The performance patterns hold steady across both Multimedia Content IO and Business IO, with the on-board NVIDIA nForce4 SATA 2 still providing the fastest IO, followed closely by the Intel ICH7R, Silicon Image 3132, and Marvell 88SE6141 SATA 2 controllers. The Silicon Image 3114 solution on the Intel D975XBX is a RAID only controller and resides on the PCI bus.

Gaming Performance Firewire and USB Performance
Comments Locked

34 Comments

View All Comments

  • ruprecht - Thursday, January 26, 2006 - link

    What the hell is wrong with Americans? Just say bad ass, it's hardly a hard-core expletive!
  • yacoub - Friday, January 27, 2006 - link

    Professional image versus immaturity. Anandtech has a professional image and likely wants to maintain it.
  • Griswold - Friday, January 27, 2006 - link

    ...land of the free. Whatever.
  • DigitalFreak - Thursday, January 26, 2006 - link

    We're prudes! Well, not me, but...
  • Avalon - Thursday, January 26, 2006 - link

    This board is pretty pitiful, IMO, when several other solutions out there can outpace it. Unless the board is offered cheaper than the competition, I see no reason to buy it.
  • JarredWalton - Thursday, January 26, 2006 - link

    Intel's proven reliability and support. Some people will happily give up 10% or more performance just for a major name like Intel. I'm not one of them, but they do exist.... :)
  • yacoub - Friday, January 27, 2006 - link

    But you pay a lot more for it...
  • Zebo - Friday, January 27, 2006 - link

    No knowledgeable computer hardware expert seriously believes that there is enough of a stability difference between the platforms to warrant a scientific comparison or research study by an objective outsider. Most "issues" stem from OE or bloated buggy SW and drivers, simple as that - not the hardware itself.

    Toms tries such a comparison but as you know a sample size of one is'nt very scientific so we'll discount intels 11 reboots two Asus boards, two intel boards, one gigabit board to AMD's 0 reboots in the stress test.:D:)
  • Zebo - Friday, January 27, 2006 - link

    Or should I say where's the proof over anything else? Sounds like urban legend to me. Something the intel boys claim with almost programed resposivness then stop replying in that thread once asked for proof.
  • Sunbird - Thursday, January 26, 2006 - link

    Excuse my ignorance, is Bad Axe a new american slang word?

Log in

Don't have an account? Sign up now