The Clear Benefits of Defragging

Posted:
in macOS edited January 2014
After reading and re-reading in forum upon forum that you don't need to defrag in Mac OSX, I decided to try it out for myself.



The difference: night and day. I defragged my computer's hard drive using Drive Genius, which only took just under an hour. Huge difference. Everything feels much more snappy. Apparently, even though OSX automatically defrags files under 20MB, defragging manually every once in a while has its benefits. Think about it: how many files under 20MB are actually haphazardly deposited on your drive? If anything, smaller files are LESS LIKELY to need defragging; large files require more space and thus have the potential to be put in a bunch of random places, slowing down your drive.



I liked these results, so I moved onto my external drive.



Big difference. I defragged my all-purpose drive (mainly larger video files, and this drive sees a lot of traffic). Transfer speeds increased, and the drive seems much faster. There was a lot of fragmentation on this drive.



I defragged my backup drive. Again, large difference; I even noticed a transfer speed increase when backing up my computer's drive to the backup while using SuperDuper. Basically it does everything twice as fast, all because I defragged the source and target drives!



The conclusion I've come to is that defragging your drive, even in OSX, is a good thing to do from time to time. Sure, there's a bit of risk involved, but the performance increase it definitely noticeable. Can anyone verify my results?

Comments

  • Reply 1 of 16
    Quote:
    Originally Posted by flinch13 View Post


    After reading and re-reading in forum upon forum that you don't need to defrag in Mac OSX, I decided to try it out for myself.



    The difference: night and day. I defragged my computer's hard drive using Drive Genius, which only took just under an hour. Huge difference. Everything feels much more snappy. Apparently, even though OSX automatically defrags files under 20MB, defragging manually every once in a while has its benefits. Think about it: how many files under 20MB are actually haphazardly deposited on your drive? If anything, smaller files are LESS LIKELY to need defragging; large files require more space and thus have the potential to be put in a bunch of random places, slowing down your drive.



    I liked these results, so I moved onto my external drive.



    Big difference. I defragged my all-purpose drive (mainly larger video files, and this drive sees a lot of traffic). Transfer speeds increased, and the drive seems much faster. There was a lot of fragmentation on this drive.



    I defragged my backup drive. Again, large difference; I even noticed a transfer speed increase when backing up my computer's drive to the backup while using SuperDuper. Basically it does everything twice as fast, all because I defragged the source and target drives!



    The conclusion I've come to is that defragging your drive, even in OSX, is a good thing to do from time to time. Sure, there's a bit of risk involved, but the performance increase it definitely noticeable. Can anyone verify my results?



    okay.
  • Reply 2 of 16
    hirohiro Posts: 2,663member
    Yes ladies and gentlemen, here we have a perfect example of the Placebo Effect!



    Or he has his drives so full that defragging actually did something. For a few minutes. And all that benefit will disappear over the next several hours of use due to drive "fullness".



    My vote is the former, especially since a manual defrag breaks the hot zone and now he has a provably longer access time to every routinely used file on his hard drive.
  • Reply 3 of 16
    irelandireland Posts: 17,798member
    @ flinch13, did you purchase that app ?
  • Reply 4 of 16
    flinch13flinch13 Posts: 228member
    Quote:
    Originally Posted by Ireland View Post


    @ flinch13, did you purchase that app ?



    No- my sister bought it, so I decided to try it out. Why not, right?



    You guys can laugh all you want, I'm convinced. Did nothing happen when you tried it, or have you never bothered to?
  • Reply 5 of 16
    mr. memr. me Posts: 3,221member
    Quote:
    Originally Posted by Hiro View Post


    Yes ladies and gentlemen, here we have a perfect example of the Placebo Effect!



    Or he has his drives so full that defragging actually did something. For a few minutes. And all that benefit will disappear over the next several hours of use due to drive "fullness".



    ...



    What you have here from the OP is an assertion. Drive Genius is not a "one trick pony." It is a general maintenance utility which includes defragmentation among its options. On a Windows computer, one would expect a much greater improvement than a two-fold speed improvement. On a Mac, would would expect a much smaller [or no] improvement from defragmentation. Deleting old files, locking-out bad blocks, or any number of other things may account for the reported speed improvement. Or, it could be the Placebo Effect.



    My bet? The OP probably freed-up enough space to reach the 10% free space threshold which allows the MacOS X virtual memory system to operate properly.
  • Reply 6 of 16
    kishankishan Posts: 732member
    Quote:
    Originally Posted by Mr. Me View Post


    My bet? The OP probably freed-up enough space to reach the 10% free space threshold which allows the MacOS X virtual memory system to operate properly.



    I can attest to the fact that this is true. As I have digitized my DVD collection my hard drive filled up quickly. At one point, my 250GB drive had only 5GB free and the Mac slowed considerably despite having 2GB of RAM. Once I off-loaded my videos onto an external drive and freed up the space, I found that the computer sped up noticeably.



    Still, do people mean to imply that drive maintenance utilities have no use on the Mac? After all, there aren't some Mac-specific drives that Apple are using. Our drives should be just as prone to developing problems as any PC's. I wish that GRC made a version of SpinRite for the Mac for the inevitable time when my drive does fail. Not that I could replace it by myself anyways...
  • Reply 7 of 16
    carniphagecarniphage Posts: 1,984member
    When new to the Mac, fresh from Windows, I tried de-fragging to gain a bit of extra snappiness TM.



    When I measured things, it was worse. The de-frag had scattered system files apart, that work better together.



    If you want to squeeze some extra speed out of your system, invest in RAM. A couple of spare gigs really does have some effect.
  • Reply 8 of 16
    kickahakickaha Posts: 8,760member
    Quote:
    Originally Posted by Carniphage View Post


    When new to the Mac, fresh from Windows, I tried de-fragging to gain a bit of extra snappiness TM.



    When I measured things, it was worse. The de-frag had scattered system files apart, that work better together.



    If you want to squeeze some extra speed out of your system, invest in RAM. A couple of spare gigs really does have some effect.



    Hear, hear.



    Kishan, while the hardware is the same, the way the OS accesses the drive, and what data it asks it to store, and how, is *very different* on Windows and MacOS X. This is what is meant by a 'filesystem'. Windows uses NTFS and FAT32, almost exclusively, while MacOS X uses HFS+ as the default. Linux systems uses ext3, or ReiserFS, or ufs, or what have you. Each filesystem approach has its pros and cons, of course, but HFS+ is pretty danged good at keeping this in line. Perfect? No. But Apple has come up with a good approach that manages the types of files that most people get fragmented badly, and are easy to fix (under 20MB), they keep track of which files are most often used and cluster them together for faster access (hot zones), etc, etc.



    Also, most drives these days come with quite a bit of logic built in that is customized to the *particular* hardware inside - not all drives are created equal. This means that many disk 'optimization' utilities actually fight *against* what the drive manufacturer has set up as the optimal use pattern, because they don't take into account (or don't know about) the internals of the drive. Some times a defrag or other optimization can make things slower because of this.
  • Reply 9 of 16
    kishankishan Posts: 732member
    Quote:
    Originally Posted by Kickaha View Post


    Hear, hear.



    Kishan, while the hardware is the same, the way the OS accesses the drive, and what data it asks it to store, and how, is *very different* on Windows and MacOS X. This is what is meant by a 'filesystem'. Windows uses NTFS and FAT32, almost exclusively, while MacOS X uses HFS+ as the default. Linux systems uses ext3, or ReiserFS, or ufs, or what have you. Each filesystem approach has its pros and cons, of course, but HFS+ is pretty danged good at keeping this in line. Perfect? No. But Apple has come up with a good approach that manages the types of files that most people get fragmented badly, and are easy to fix (under 20MB), they keep track of which files are most often used and cluster them together for faster access (hot zones), etc, etc.



    Also, most drives these days come with quite a bit of logic built in that is customized to the *particular* hardware inside - not all drives are created equal. This means that many disk 'optimization' utilities actually fight *against* what the drive manufacturer has set up as the optimal use pattern, because they don't take into account (or don't know about) the internals of the drive. Some times a defrag or other optimization can make things slower because of this.



    Appreciate the education... When I have some time, I'll be browsing the wikipedia entries for the various file systems. Do you happen to know if the way NFTS vs FAT32 vs HFS deal with data makes any difference in the long term viability of the hardware? Or is this more dependent on things like heat, movement of the drive etc?
  • Reply 10 of 16
    kickahakickaha Posts: 8,760member
    Much more dependent on the physics (and manufacturing quality) of the drive.



    One thing I didn't say is that while the average usage pattern is well served by Apple's default setup of HFS+, some folks don't meet that pattern.



    Folks doing real-time video editing of massive files, for instance, *do* need to keep their working area drive space as contiguous (defragged) as humanly possible, contingent on the layout algorithms of the drive itself. These files are well outside the 20MB limit, and sorely need constant continuous data access. For them, a defrag solution for the drive they store their media files on is a must - but the folks who are going to be working on 300GB files are pros who already *know* this, so they'll take their own steps to take care of it.



    It's one of those things: if you have to ask, you don't need it.
  • Reply 11 of 16
    mydomydo Posts: 1,888member
    We should do a randomized study.



    Frag and defrag a drive and have the user rate the computer performance on the snappy meter.
  • Reply 12 of 16
    kickahakickaha Posts: 8,760member
    Much better with a stopwatch...
  • Reply 13 of 16
    dentondenton Posts: 725member
    Quote:
    Originally Posted by Kickaha View Post


    Much better with a stopwatch...



    Oh, come on, don't you think the "snappy meter" sounds more fun?
  • Reply 14 of 16
    Quote:
    Originally Posted by Denton View Post


    Oh, come on, don't you think the "snappy meter" sounds more fun?



    it's called a snapometer. you could use mine but the batteries in it are dead and my last fresh set are in the chancesofgettinglaidometer.
  • Reply 15 of 16
    hirohiro Posts: 2,663member
    Quote:
    Originally Posted by ghiangelo View Post


    it's called a snapometer. you could use mine but the batteries in it are dead and my last fresh set are in the chancesofgettinglaidometer.



    Let me borrow that one, but put the dead batteries in first, it will read the same...
  • Reply 16 of 16
    mydomydo Posts: 1,888member
    Quote:
    Originally Posted by ghiangelo View Post


    it's called a snapometer. you could use mine but the batteries in it are dead and my last fresh set are in the chancesofgettinglaidometer.



    Yes the snapometer. Measured in units of Steves (Sv). If your batteries are dead you can send it to Apple and get them replaced for only $96.
Sign In or Register to comment.