Are you sure about this, Chuck? Seems to me that a number of practices in
current systems could lead to severe fragmentation. One is "write before
erase", which I understand is often used in writing out changes to files
while those files are still open for editing, so that things like immediate
undo are possible. Another is "sparse file writes", which, rather than
copying the entire file to a new location on save, just writes out the
changes and additions with appropriate pointers in the old file. And the
third I can think of is the practice that I know used to be true, where in
order to speed up disk write times Microsoft systems used to use what
amounted to "nearest large enough area" on write. That meant that "holes"
left by having re-written files to a larger free area when they grew were
often left behind, and then either only partially filled by the next
optimistic write, or opportunistically used on the next write of a file
that would be too large for the space, resulting in the file being
fragmented across multiple extents.
---
Scott Gomez
On Tue, Aug 14, 2012 at 5:01 PM, Chuck Norcutt <
chucknorcutt@xxxxxxxxxxxxxxxx> wrote:
> But even with FAT32 defrag will never make significant performance gains
> unless the space has previously been completely filled. If the system
> has a large drive that's never been filled up defrag has nothing to do.
>
> Chuck Norcutt
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|