Pure Hard Disk Performance - IPEAK

We begin our usual hard disk drive test session with Intel's IPEAK benchmarking utility. We first run a trace capture on Winstone 2004's Business and Multimedia Content Creation benchmark runs to catch all of the IO operations that take place during each test. We then play back each capture using RankDisk, which reports back to us a mean service time, or average time that the drive takes to complete an IO operation.

IPEAK Business Winstone 2004 - Pure Hard Disk Performance

The first batch of 3.0Gbps hard drives that we tested gave us results, which were scattered all over the charts for the IPEAK tests with Western Digital's WD1600JS, taking the majority of the wins over the other two 3.0GBps drives. This time around, Seagate's 500GB 7200.9 takes place half way down the list at 541 IO operations per second.

Let's take a look at Content Creation performance.

IPEAK Content Creation Winstone 2004 - Pure Hard Disk Performance

The Seagate drive jumps to the 3rd spot on the list for the Content Creation Winstone IPEAK run at 389 IO operations per second.

IPEAK Average Read Service Time

As far as Read Service times go, Seagate's drive reports 13.9ms on average, which is just a fraction of a millisecond slower than the WD1600JS, but more than 2ms quicker than the HD160JJ.

The Test Overall System Performance - Winstone 2004
POST A COMMENT

46 Comments

View All Comments

  • gutharius - Wednesday, August 30, 2006 - link

    Wanted to ask if during the benchmarking your removed the jumper that sets the Barracuda to SATA150 as opposed to SATA300? Seagate manufactures this drive with the drive set to run SATA150, for compatibility reasons with the VIA chipset. Please check this as I think this drive would fair much better if it were set to run in SATA300 not SATA150. Reply
  • ROcHE - Friday, February 10, 2006 - link

    This is getting confusing with all those platters size. Would be great to have these compaired. Reply
  • kd4yum - Thursday, October 27, 2005 - link

    Hey
    The whole NCQ process has its 'glamour' from SCSI. But really, SCSI just brought it down from Mainframe tech. IF you got 100+ people working a DL/1 or DB2 database, then 'Elevator' queuing pays. Maybe also in a file server (SCSI). But forget it in a Personal Computer! Who you gonna queue?
    Its like trying to do file compression in software during a backup. You lose more in computation than you gain in I/O.
    sorry for the flame
    Reply
  • eastvillager - Wednesday, October 26, 2005 - link

    You know how your mouth starts to water right before you're about to vomit? :-) Reply
  • Peter - Tuesday, October 25, 2005 - link

    yeah right ... I was so bored with those results I started drooling. Reply
  • formulav8 - Monday, October 24, 2005 - link

    I really do not care for the style of this article/review from this new guy. (I am guessing he is new? Don't reconize the name?)

    I can definitely tell if Anand himself his the author of the article/review or if its one of his fill-in guys. But they do seem to be getting better with time.

    And as many others has said, the title of this article is completely 'false' since the review clearly shows the drive to be average at most. This drive is dissappointing for the features it has. High density platters and 16MB cache and it still gets beat by the older 8MB Cache/Lower Density platter drives. Unless I am looking at the graphs wrong??


    Jason
    Reply
  • Scrogneugneu - Monday, October 24, 2005 - link

    quote:

    Unless I am looking at the graphs wrong??


    Well, look at the graphs the way you want, the middle will always stay in the middle...
    Reply
  • tomoyo - Monday, October 24, 2005 - link

    Well first of all, I fully agree that the title "mouthwatering" makes no sense whatsoever. Even the conclusion is so-so at best...if the seagate drive had beaten all the other 7200rpm drives..that would be true mouthwatering numbers. It's not even a winner in terms of storage since there's already 500gb drives out there. Nor does it advance platter sizes like the single platter design. So that title is completely false in my eyes.

    Secondly, I've noticed the writing and analysis has gone greatly downhill since Anand himself hasn't been writing as much. Anand tends to have some insightful commentary on the new technology involved and explainations of the numbers. There's at best shallow explanations of numbers and nothing really distinguishing what's really happening with the benchmarks. And as noted, there's nothing that explains all the different platter sizes in any depth, nor into what pluses or minuses the full SATA 2.5 interface might have in terms of benchmarks and usability. I have to say this is a very disappointing overall job, and I hope the writing and analysis increase by a magnitude of at least 10 to keep up to anandtech standards.
    Reply
  • Veerappan - Monday, October 24, 2005 - link

    On the final words page is a pic of a read speed test that was performed on this drive.

    http://images.anandtech.com/reviews/storage/seagat...">http://images.anandtech.com/reviews/sto...eagate_b...

    Maybe I am seeing this wrong (or it could be a program limitation), but why is this program reporting that the theoretical limit was 150MB/s? If this drive was being tested at 3.0Gb/s, shouldn't that figure be roughly double what it is?
    Reply
  • kd4yum - Thursday, October 27, 2005 - link

    Yeah
    Reply

Log in

Don't have an account? Sign up now