outer edges of a hard drive are NOT the best place

Discussion in 'Windows Desktop Systems' started by Perris Calderon, Sep 22, 2004.

  1. Perris Calderon

    Perris Calderon Moderator Staff Member Political User

    Messages:
    12,332
    Location:
    new york
    ha,,,just found out another of those tech folk lore that everybody beleives, but isn't actually true...man I love these;

    on the current hard drives, seek time actually turns out to be worse not better on the first and last few percent of the drive's cylinders.

    get this...drive software in some cases will slow the arm down in the outer edges so it doesn't hit the limit stops...obviously,hitting those is a vewy vewy bad.

    I LOVE when and old notion is no longer true...*waits for the proper moment to spring this info on someone and start a huge debate*

    anyway, it's just a small percentage of those cylendars that get slowed down, I suspect the first protion of a hard drive is still the optimum portion of the drive if it's going to be an average sized partition
     
  2. NetRyder

    NetRyder Tech Junkie Folding Team

    Messages:
    13,256
    Location:
    New York City
    This is something I always believed to be true. [​IMG]
    The interesting thing is that PerfectDisk places the most frequently accessed/modified files at the center of the drive, but places the prefetch folder contents at the starting edge of the drive. I wonder why they chose to do that. Any ideas?
     
  3. Perris Calderon

    Perris Calderon Moderator Staff Member Political User

    Messages:
    12,332
    Location:
    new york
    cause while the os is booting, nothing but those files are important, and in that case, the beginning of the drive would be the best spot...once the operating system has boot and other files are getting accessed, the middle of frequently accessed files would be where the heads will be the majority of the time, and that's where the most optimum spot on the disc would be

    but before the volume is mounted, the outer edge, would do it...because there is no seek time on boot, the heads would just start there and stay there through the boot process
     
  4. NetRyder

    NetRyder Tech Junkie Folding Team

    Messages:
    13,256
    Location:
    New York City
    Makes sense. [​IMG]
     
  5. Petros

    Petros Thief IV

    Messages:
    3,038
    Location:
    Pacific Northwest
    Yay perris! Ignorance is an ugly thing. Well, maybe just not keeping up with newer technologies is what it is. Either way, good info! Thanks.
     
  6. Perris Calderon

    Perris Calderon Moderator Staff Member Political User

    Messages:
    12,332
    Location:
    new york
    ya, I just had the discussion with Jaimie Hanrahan, and I posed your question to him...he confirms my thoughts, but adds some very interesting info;

    Prefetch actually arranges fragments of those files, (not the entire file) in the order in which the OS will most likely, (using statistic algorithms), read them... generally, these files aren't read from beginning to end before the head goes to the next file...prefetch lets the disk head behave as if they are being read beginning to end...for this, the heads move very little at boot...cute

    here's another cute ditty he told us;

    Seagate used to ship a hard drive called the "WrenRunner".

    more expensive, with 10% less capacity over their other hardrive.

    get this...it had average access times 20% better then their other model, and so they charged more for the hardrive!

    HA!!! all Seagate did was lock out the first and last 5% cylinders!!!!

    Now with the stop blocks safely far away from the cylinders that would actually be used they can get full seek speeds across the entire drive!...obviously, reducing the maximum stroke distance contributed as another benefit.

    So Jaimie used to used to tell his clients to buy the cheaper/larger/"slower" drive, and just not use all of it. Same principle here.

    my feelings are though that Seagate also sped the seek arm for this particular drive, since the stops weren't an issue anymore...so simply not using the first few cylinders would in my mind not quite emulate the affect.

    anyway, a great story
     
  7. NetRyder

    NetRyder Tech Junkie Folding Team

    Messages:
    13,256
    Location:
    New York City
    Very cool, perris. [​IMG]
     
  8. VenomXt

    VenomXt Blame me for the RAZR's Folding Team

    Messages:
    3,453
    Location:
    Houston, Texas
    good read peris thanks.