On Thu, 23 Jul 2009 15:34:41 -0700, Frank wrote:
[color=blue]
> ray wrote:[color=green]
>> On Thu, 23 Jul 2009 13:14:31 -0700, Frank wrote:
>>[color=darkred]
>>> ray wrote:
>>>> On Thu, 23 Jul 2009 09:36:22 -0700, Frank wrote:
>>>>
>>>>> +Bob+ wrote:
>>>>>> On 23 Jul 2009 15:22:26 GMT, ray <ray@zianet.com> wrote:
>>>>>>
>>>>>>> I most certainly DO have it. And I did not mention Ubuntu until
>>>>>>> YOU brought it up. I was merely making the point that in the 21st
>>>>>>> century a properly designed file system does not need continual
>>>>>>> defragging to maintain responsive performance. Please explain how
>>>>>>> that is 'pushing' Ubuntu.
>>>>>> MS started claiming when NT came out that NTFS did not need to be
>>>>>> defragged. Actual performance measurement proved them incorrect :-)
>>>>> Yeah, just like linux claims it never needs to be defragged...that's
>>>>> why there are dozens of linux defragmenting software utilities
>>>>> available!...LOL!
>>>> There are not dozens. There are a few. I've never known anyone to use
>>>> them.
>>> There are more than "a few" and their very existence belies your
>>> assertion that they are not need.
>>> The FUD you spread is noted.[/color]
>>
>> Yes frankie they exist. If you do a web search you might actually be
>> able to find some. The ones of which I am aware defragment ext2 file
>> systems. I've never heard of one for a more modern fs such as reiser.
>> You will also find that NO SIGNIFICANT Linux distribution includes a
>> defragmentation tool in it's base installation. It must not be very
>> necessary.
>>
>> Just for grins, do a web search for 'defragmentation Linux' - most of
>> what you will find are explnations of why Linux does not need it.[/color]
>
> Just for grins..read more...there is some very telling evidence that
> ext3 needs to be defragged occasionally![/color]

And ext3 is merely ext2 with journaling. As I said, none for modern fs
such as reiser et. al.

In message <8tnh65deiqbre7alkdjqp09csoaj2lip51@4ax.com> +Bob+
<nomailplease@example.com> was claimed to have wrote:
[color=blue]
>On Thu, 23 Jul 2009 14:46:25 -0700, Dave Warren
><dave-usenet@djwcomputers.com> wrote:
>[color=green]
>>In message <6nof65ti1jnm5iddj8oeaqm3h6otitfq6g@4ax.com> +Bob+
>><nomailplease@example.com> was claimed to have wrote:
>>[color=darkred]
>>>Background defrag is a waste of performance you could be using in the
>>>foreground.[/color]
>>
>>Vista's defragmentation tool only works while the system is idle, both
>>by way of a backoff algorithm, and also the disk queuing system.
>>[color=darkred]
>>>Not to mention, defragging a disk is a moderately risky
>>>operation on the drive. Doing it while foreground tasks are running is
>>>not very smart.[/color]
>>
>>There is virtually no risk at all with a modern defragmentation tool
>>using Windows' NTFS defragmation APIs. When a cluster is moved, there
>>are several steps, each of which is journaled so at no point will a
>>failure or interruption cause corruption or data loss.[/color]
>
>If there's a power failure or crash, you can easily end up with
>errors.[/color]

Look up "journaling" -- A properly journaled filesystem implementation
won't result in any corruption after a failure.

The short version is that whenever the filesystem's structures are being
updated, the OS makes a note about what changes it's about to do in the
journal, sets a dirty flag, makes the change, then updates the journal
and dirty flag to indicate the changes are completed.

In the event of a failure (power, crash, etc), when the OS comes back up
it sees the "dirty" flag and reviews the journal to complete the
changes.

On Thu, 23 Jul 2009 00:14:24 -0400, +Bob+ <nomailplease@example.com>
wrote:
[color=blue]
>On Wed, 22 Jul 2009 21:04:39 -0700, Ashton Crusher <demi@moore.net>
>wrote:
>[color=green]
>>I understand your desire to watch colors change on the screen but the
>>built in defrager works fine if you use it as defraging is intended to
>>be used - as an out of sight/out of mind background task. I set it up
>>that way 18 months ago and I have never even noticed it running since
>>then.[/color]
>
>Background defrag is a waste of performance you could be using in the
>foreground. Not to mention, defragging a disk is a moderately risky
>operation on the drive. Doing it while foreground tasks are running is
>not very smart.
>[/color]

Well, if you think that's a risk worth worrying about then no one
should ever install any programs on their computer since program
installs trash far more systems then a background defrager ever has.
If you run the built in defrag as it's intended it simply is always
doing little bits of clean up and there is no need for it to take over
the system like the store bought ones do so that people can look at
the pretty little squares changing colors - been there, did that, am
over it.

[color=blue]
>Just get a decent defrag utility and run it periodically when you
>won't be doing anything on the system. Shut off Vista's useless dumbed
>down defrag. All fixed.[/color]

"+Bob+" <nomailplease@example.com> wrote in message
news:vi2h65p6de5piekrlmv8n0hqmrhuvq7t6e@4ax.com...[color=blue]
> On Thu, 23 Jul 2009 12:27:25 +0100, "Gordon" <gordonbparker******.com>
> wrote:
>[color=green][color=darkred]
>>> And it does slow down after time.[/color]
>>
>>But that is NOT, in general, due to a fragmented HDD!
>>Defragging makes very little difference, in general, to system
>>performance.[/color]
>
> Inaccurate. On some systems it makes a tremendous difference. It
> depends entirely on how much activity (add/delete) the disk drive
> sees.
>
>[/color]

Not inaccurate at all. In general, most users will notice only very slight
improvements by defragging.
Speed depends on many factors of which an unfragged HDD is a very small
one...

"+Bob+" <nomailplease@example.com> wrote in message
news:dg2h65ha1j9ivremnohctkjs83m5bs8fre@4ax.com...[color=blue]
> On Thu, 23 Jul 2009 07:42:48 +0100, "Gordon" <gordonbparker******.com>
> wrote:
>[color=green]
>>
>>The Vista defrag only kicks in when the system is IDLE......so no loss of
>>performance there then, is there?[/color]
>
> Yes there is. It's a noticeable thrashing of the drive. When you start
> a foreground program and the disk is being thrashed by the defrag, you
> can see a noticeable delay.[/color]

Rubbish. Never noticed that AT ALL.

[color=blue]
>
> Not to mention, it's notoriously inefficient. Go to any system running
> the background defrag. Run a defrag on it. Notice how long it has to
> run, even though it's supposedly defragged.
>[/color]

<sigh> WHY? WHY even BOTHER to run a manual defrag if it's doing it FOR you
automatically? If it ain't broke DON'T FIX IT!!!!!!!

The Vista defragger is painfully slow and rather inefficient. It also
lacks a GUI, drive map, progress bar etc. If you are interested in an
advanced defragmenter that has both fully automatic as well as manual
defrag modes with GUI, drive map...then Diskeeper is the best choice.

"GenFil" <GenFil.3vtwxp@no.email.invalid> wrote in message
news:GenFil.3vtwxp@no.email.invalid...[color=blue]
>
> The Vista defragger is painfully slow[/color]

How do you know? It works in the BACKGROUND.....
[color=blue]
> and rather inefficient. It also lacks a GUI,[/color]

So how do you know it's inefficient if you can't SEE it?

<sigh>

On Thu, 23 Jul 2009 21:23:35 -0700, Dave Warren
<dave-usenet@djwcomputers.com> wrote:
[color=blue]
>
>Look up "journaling" -- A properly journaled filesystem implementation
>won't result in any corruption after a failure.
>
>The short version is that whenever the filesystem's structures are being
>updated, the OS makes a note about what changes it's about to do in the
>journal, sets a dirty flag, makes the change, then updates the journal
>and dirty flag to indicate the changes are completed.
>
>In the event of a failure (power, crash, etc), when the OS comes back up
>it sees the "dirty" flag and reviews the journal to complete the
>changes.[/color]

That's fine for transaction logging in applications when the disk
drive and OS is considered a known stable entity. When done at the
root level of the disk drive with all of the issues of an OS having to
track the changes it's not dependable as you lost two of the given
reliability factors (disk drive and OS).

Manfred Nathanal Hollingsworth wrote:[color=blue]
> Gordon wrote:[color=green]
>>
>> "+Bob+" <nomailplease@example.com> wrote in message
>> news:vi2h65p6de5piekrlmv8n0hqmrhuvq7t6e@4ax.com...[color=darkred]
>>> On Thu, 23 Jul 2009 12:27:25 +0100, "Gordon" <gordonbparker******.com>
>>> wrote:
>>>
>>>>> And it does slow down after time.
>>>>
>>>> But that is NOT, in general, due to a fragmented HDD!
>>>> Defragging makes very little difference, in general, to system
>>>> performance.
>>>
>>> Inaccurate. On some systems it makes a tremendous difference. It
>>> depends entirely on how much activity (add/delete) the disk drive
>>> sees.
>>>
>>>[/color]
>>
>> Not inaccurate at all. In general, most users will notice only very
>> slight improvements by defragging.
>> Speed depends on many factors of which an unfragged HDD is a very
>> small one...[/color]
>
> Err <sigh> you don't know what most users will notice unless Err <sigh>
> you spoke with most users. Err <sigh>
>[/color]


you are sighing.. frank must be giving you his bone...

No i dont mean his microscopic dick, rather a real bone...

got it bonehead?

In message <qifj65havn3pn2hqulahkqrg2riab40647@4ax.com> +Bob+
<nomailplease@example.com> was claimed to have wrote:
[color=blue]
>On Thu, 23 Jul 2009 21:23:35 -0700, Dave Warren
><dave-usenet@djwcomputers.com> wrote:
>[color=green]
>>
>>Look up "journaling" -- A properly journaled filesystem implementation
>>won't result in any corruption after a failure.
>>
>>The short version is that whenever the filesystem's structures are being
>>updated, the OS makes a note about what changes it's about to do in the
>>journal, sets a dirty flag, makes the change, then updates the journal
>>and dirty flag to indicate the changes are completed.
>>
>>In the event of a failure (power, crash, etc), when the OS comes back up
>>it sees the "dirty" flag and reviews the journal to complete the
>>changes.[/color]
>
>That's fine for transaction logging in applications when the disk
>drive and OS is considered a known stable entity. When done at the
>root level of the disk drive with all of the issues of an OS having to
>track the changes it's not dependable as you lost two of the given
>reliability factors (disk drive and OS).[/color]

If your drive(s) is(/are) failing in such a way that they report success
but actually fail, you've got bigger issues then a defrag.

As far as the OS side of things, can you point to one cite anywhere in
the history of Vista (or XP, for that matter) showing a flaw in the NTFS
defrag APIs?

The only gap that you can encounter is drive controllers that do
hardware caching but fail to flush the drive on demand -- Software can't
compensate for broken hardware, nor should it be expected to try (and
the same goes for OSes attempting to compensate for broken drivers)

"Ashton Crusher" <demi@moore.net> wrote in message
news:64ei65l7qs1f5jla6kai3apus98k30n05u@4ax.com...[color=blue]
> On Thu, 23 Jul 2009 00:14:24 -0400, +Bob+ <nomailplease@example.com>
> wrote:
>[color=green]
>>On Wed, 22 Jul 2009 21:04:39 -0700, Ashton Crusher <demi@moore.net>
>>wrote:
>>[color=darkred]
>>>I understand your desire to watch colors change on the screen but
>>>the
>>>built in defrager works fine if you use it as defraging is intended
>>>to
>>>be used - as an out of sight/out of mind background task. I set it
>>>up
>>>that way 18 months ago and I have never even noticed it running
>>>since then.[/color]
>>
>>Background defrag is a waste of performance you could be using in
>>the
>>foreground. Not to mention, defragging a disk is a moderately risky
>>operation on the drive. Doing it while foreground tasks are running
>>is
>>not very smart.[/color]
>
> Well, if you think that's a risk worth worrying about then no one
> should ever install any programs on their computer since program
> installs trash far more systems then a background defrager ever has.
> If you run the built in defrag as it's intended it simply is always
> doing little bits of clean up and there is no need for it to take
> over
> the system like the store bought ones do so that people can look at
> the pretty little squares changing colors - been there, did that, am
> over it.[/color]

Practically speaking, just running a defrag on a set schedule and
preferably when you won't be using it is the best advice. I do that
with my various back-ups and security sweeps.

However, there's more information in these other defrag programs than
watching the lights blink, such as listing files that won't defrag.
When hiberfil.sys came up on a manual defrag of my laptop as a file
that could not be defragged, I deleted it and recovered quite a bit of
space. I had hibernation turned off but didn't realize that it
already had created a whopping useless file. Deleting it also allowed
the defragger to pack more files into the first band of written
clusters. I'm sure the performance increase was trivial, but the
recovered space on a relatively small laptop HD was worth doing.

Seeing the list of undefraggable files also taught me that in order to
squeeze the absolute most out of defragging, shut down every (and I do
mean every) other app before defragging. An open program can put a
hold on auxiliary files it creates while running and the defragger
won't touch them. Again, the effect is small but if you're not using
the computer, why not shut down everything?

Perusing the other undefraggable/unmovable files and the pretty
squares also can be instructive as to what's happening "under the
hood". For example, most writes are done in two bands, one starting
at the beginning of the HD, and another band starting at about the
middle of the drive. This must be a deliberate write strategy.
Otherwise, used clusters would be randomly strewn across the drive.
I've also backtracked undefraggable/unmovable files to one or two
resource-consuming processes that I didn't need and could be
terminated.

I find perfect disk to be the best at defragging. I love it and use it all
the time.


"ColTom2" <noemailaddress@nomail.com> wrote in message
news:eTlTj8iCKHA.3556@TK2MSFTNGP04.phx.gbl...[color=blue]
> Hi:
>
> Windows Vista Defragmenter is about the worst that I have ever seen. It
> does not show you any progression status and it takes forever.
>
> Does anyone have an alternate free defragmenter and/or suggestion?
>
> Thanks,
>
> ColTom2
>
>[/color]

--
Steven J. Hayes

"Steve H." <vze245by@verizon.net> wrote in message
news:871CF9BD-B847-465C-BCE0-6AD4E6A5094D@microsoft.com...[color=blue]
> I find perfect disk to be the best at defragging. I love it and use it all
> the time.
>[/color]

<sigh>

"Bill Daggett" <Daggett@wmunny.invalid> wrote in message
news:ju5f65tju4c2l8oqq361bqb5brlnrpva6r@4ax.com...[color=blue]
> Bill Sharpe <wfsnopam@adelphia.net> wrote:
>[color=green][color=darkred]
>>> A fragmented HDD is one of the least likely causes of a slow machine....[/color]
>>
>>True enough, but if defrag will slow down the machine while it's
>>running. Since defrag seems to work in the background it's not always
>>easy to tell that it is running.[/color]
>
> Why worry about it? Defragging is over-rated anyway.[/color]

Defragging must be done, once you run out of monsters to frag in your
favorite FPS.

"JamesJ" <jjy@darwin_roadrunner.com> wrote in message
news:%23NHbaO5CKHA.4004@TK2MSFTNGP05.phx.gbl...[color=blue]
>I don't worry about it. I just do those things for myself.
> I defrag, maybe once a month. I just feel the more items running
> slows my system down. And since I've been tweaking Vista services
> I've noticed my system starting faster and not having to wait for 5
> minutes
> after the desktop shows up before I can do anything.
>
> James
>[/color]
You need to check your startup programs, from your startup folder and
registry startup keys, if you're waiting 5 minutes before you can do
anything.
I timed my Vista system from completely powered off to loaded a web page in
IE in under 90 seconds.
My XP system does the same within 3 minutes, including time it takes to
process group policy settings.
I defrag my XP system once a year, and I do a LOT of file processing.
That's not installing/uninstalling some shareware products. That's
routinely copying/deleting 1.5GB worth of data files at a time.