about fragmentation (of free disk space)

Brandon Allbery allbery.b at gmail.com
Sat Oct 10 12:09:24 PDT 2015


On Sat, Oct 10, 2015 at 2:45 PM, Daniel J. Luke <dluke at geeklair.net> wrote:

> On Oct 10, 2015, at 7:28 AM, René J.V. Bertin <rjvbertin at gmail.com> wrote:
> > HFS+ is supposed to contain algorithms that limit file fragmentation,
> but without a background process that moves files (or file blocks), it
> cannot prevent free space fragmentation, just limit it. On a spinning disks
> that can become a limit on performance (I presume that theoretically the
> same applies to SSDs too)
>
> random access time for a SSD is 1-3 orders of magnitude less than for a
> rotational drive.
>
> As with anything, you need to measure ‘real world use’ to be certain, but
> it’s probably not an issue for SSDs at all.


Additionally, we are at the point with "spinning rust" that even new disks
already have "bad blocks" because we've pushed the storage density past
what can reliably be manufactured 100% of the time --- and block remapping
is done transparently at the controller level, so you have no way of ever
finding out if your "contiguous" file really is. SSDs take this to another
level, as they move blocks around to even out writes; you can assume they
are pretty much never contiguous after the first write to the drive. (But
they're actually RAM, so this does not matter; it takes the same time to
access any block.)

tl;dr: for practical purposes, there is no such thing as "contiguous" any
more

-- 
brandon s allbery kf8nh                               sine nomine associates
allbery.b at gmail.com                                  ballbery at sinenomine.net
unix, openafs, kerberos, infrastructure, xmonad        http://sinenomine.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.macosforge.org/pipermail/macports-users/attachments/20151010/48e6df38/attachment.html>


More information about the macports-users mailing list