Pure Hard Disk Performance

To measure "pure" hard disk performance, we took a real world benchmark - in this case, the entire Winstone 2004 suite - and used Intel's IPEAK utility to capture a trace file of all of the IO operations that take place during a single run of Business Winstone 2004 and MCC Winstone 2004. We then use IPEAK to play back the trace, much like a timedemo, on each of the hard drives, which gives us 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.

Keep in mind that these performance scores are best only for comparing pure hard disk performance, and in no way do they reflect the actual real world performance impact of these hard drives.

For descriptions of what the Business and Multimedia Content Creation Winstone 2004 tests consist of, reference those benchmark pages.

IPEAK Business Winstone 2004 - Pure Hard Disk Performance

While a score of 901 IO operations per second seems impressive, what's important to note is that two Raptors in RAID-0 only offers 20% more performance than a single Raptor in this purely disk-bound test. We have already seen how aggressive caching, pre-fetching and other such real world elements of modern day computers reduce the impact of hard drive performance on overall system performance. With only a 20% improvement in pure disk performance, this doesn't bode too well for RAID-0 offering much of a real world performance boost.

IPEAK Content Creation Winstone 2004 - Pure Hard Disk Performance

With data from audio editing as well as other more IO intensive tasks, our IPEAK recording of Content Creation Winstone 2004 shows much more potential for RAID-0, with the Raptor RAID-0 array outperforming a single drive by 38%.

Now, it's time to see how these numbers translate into real world performance.

The Test Overall System Performance – Winstone
Comments Locked

127 Comments

View All Comments

  • Runamile - Thursday, July 1, 2004 - link

    I liked the diagrams for RAID0 and 1. Would be cool to see 3,4,5, and 10 drawn out too, but that wouldn't of been relevent to the article.
  • ciwell - Thursday, July 1, 2004 - link

    Excellent article...and for those who think it is faster experientially: it is all in your head. ;)
  • SilverBack - Thursday, July 1, 2004 - link

    I'm using two RAID 0 arrays.
    A8V mobo with a promise 378 controller and the onboard VIA as well.

    I prefer the system this way. It just makes the whole windows experience faster.

  • RebolMan - Thursday, July 1, 2004 - link

    Any subjective comments on whether the system using RAID-0 feels any smoother? A lot of people comment that P4s with Hyperthreading produce a system that just feels more responsive regardless of whether it's really any faster.

    I find the best thing to do (under Windows) when you've got two drives hooked up is to move your Virtual memory onto the one which you use less. There's all sorts of tricks you can use to distribute your system load without necessarilly using RAID.
  • wanosd - Wednesday, August 4, 2010 - link

    I recently did a test by copying a few GB of data from a WD 160 GB drive to another WD 160 GB drive. It took about 4 mins.

    I then renamed the folder that I just copied and then copied it back to the original drive, and again got about the same time, with only a few seconds difference.

    I timed my boot from Windows from the time the OS takes over, all the way to the desktop, and it took about 35 seconds. I do NOT have any bloatware or junk on my system.

    Finally, I enabled RAID 0 for these two drives. Now the same version of Windows boots up in about 25 seconds (not as fast as you'd think). Also, copying the same folder from my 3rd hard drive to my RAID 0 drives is taking 1 minute and 45 seconds. The seek time itself may be still slow, but once you get the data going, it'll definitely help out.
  • qepsilonp - Monday, October 31, 2011 - link

    I play eve online and i can be running at any given time 3 clients of eve a music player / video / or a text to speech program a browsing client with usually 5 - 7 tabs and sometimes i even want to be able to extract files at the same time I think for that kind of usage RAID 0 would be very worth it, did you even consider a lot of users do multiple demanding tasks at once?
  • qepsilonp - Saturday, November 5, 2011 - link

    while yes when running one application RAID 0 is usually useless but when like me most of the time you are using 2-3 clients of a heavily HDD relent game where sometimes it takes a while to get the files for the 3D images and sometimes because of that your they wont show up on the screen for sometimes 5 seconds and i know its not anything else but the HDD because i have a new computer and the only piece of hardware that hasn't been updated is the HDD and im still getting it.

    if i was able i would also be running a HD movie or have my computer read a book to me with a text to speech program or be playing music and also maybe extracting something with Winrar you cant tell me that with all those IO's that RAID 0 wouldn't help at all, considering the game im playing is eve online and when i jump in and a gate can have 1000 ships on it thats maybe 32 different ships types which have to be gotten from the HD which is probably something like 100mb times 2 - 3 thats 64 to 96 IO's thats if theres isnt multiple files that need to be called up for 1 ship type so yeah I think in the deck top for power users there is a place for RAID 0

Log in

Don't have an account? Sign up now