by jamvaru on Tue Jan 08, 2013 3:03 pm
you might consider that the benefits are negligible or non-existant
the arm takes time to move to the 'front' and 'back' of the drive (your random access time)
placing the drivemap somewhere in the middle, maybe 1/3 or so from the outer rim, means that overall the arm moves less, increasing rat
defragmenting files means taking the bits and pieces that exist on different parts of the drive and putting them in one spot in the proper order
the idea is that the computer can then read the whole file all at once, with one movement of the arm, rather than multiple movements, if it is smart enought to do so, and I believe that is why microsoft make the defrag api, because their os is capable of reading a file at once if it is contiguous
placing a file at the 'front' of the drive is considered superior (faster) because the speed of the drive is greater, so more data can be placed on a single cylinder near the front of the drive, resulting in less movement of the arm
it may also be (im just not positive, but i think so) that the data at the front is retrieved or written faster than data at the back, or near the center, regardless of it being on one or more cylinders
so...
linux has taken the position that fragmentation and optimization is unnecessary; this is true with ssd drives
supposedly linux manages fragmentation and optimization on the fly, however, it certainly fails to do so properly on ntfs partitions, probably because the authors of the code simply don't care or actively hate ntfs, even though they write the code for it; it seems they just want to mess up or ntfs partitions on purpose
at any rate, there are several questions left unanswered
1. is there a way to defrag and optimize ntfs partitions from within linux? (using a linux app or a wine app)
2. is there a way to convince linux to not screw it up from the start, so the defragger is not needed?
3. regarding $security_descriptor(s) ... how can we manipulate these from within linux, to consolidate them into one location or whatever works
4. why the lackluster response?
----
I was able to get a log of which files are unmovable, according to mydefrag, then copy the parent folder to a different drive; after then deleting the original, the 'red dots' were gone (though more remain, from a different folder... same procedure)