Roy Schestowitz <newsgroups@xxxxxxxxxxxxxxx> wrote in
news:1218062.m9IdtGTYQt@xxxxxxxxxxxxxxx:
> ____/ Peter Köhlmann on Wednesday 12 September 2007 06:58 : \____
>
>> Tim Smith wrote:
>>
>>> In article <1290252.J6yoQHQMNs@xxxxxxxxxxxxxxx>,
>>> Roy Schestowitz <newsgroups@xxxxxxxxxxxxxxx> non sequitured:
>>>> FS fragmentation -- You think that's a good thing? So just because
>>>> I save
>>>> files, a hard drive is torn to piece? How annoying. Yes, i'm sure
>>>> it CAN'T be turned off, but still... what a fucking stupid idea.
>>>
>>> Too bad Linux filesystems get fragmented, then. Try this, as root:
>>>
>>> for i in 1 2 4 8 16 32 64 128 256
>>> do
>>> dd if=/dev/zero of=$i.meg bs=1024k count=$i
>>> done
>>> filefrag *.meg
>>>
>>> (Root is needed for filefrag)
>>>
>>> That will create 9 files, ranging from 1 meg to 256 meg, and then
>>> tell you how many fragments each consists of. Here's what I got on
>>> one of my Ubuntu systems:
>>>
>>> 128.meg: 39 extents found, perfection would be 2 extents
>>> 16.meg: 14 extents found, perfection would be 1 extent
>>> 1.meg: 133 extents found, perfection would be 1 extent
>>> 256.meg: 49 extents found, perfection would be 3 extents
>>> 2.meg: 153 extents found, perfection would be 1 extent
>>> 32.meg: 8 extents found, perfection would be 1 extent
>>> 4.meg: 103 extents found, perfection would be 1 extent
>>> 64.meg: 12 extents found, perfection would be 1 extent
>>> 8.meg: 6 extents found, perfection would be 1 extent
>>>
>>> and here is another:
>>>
>>> 128.meg: 72 extents found, perfection would be 2 extents
>>> 16.meg: 47 extents found, perfection would be 1 extent
>>> 1.meg: 20 extents found, perfection would be 1 extent
>>> 256.meg: 50 extents found, perfection would be 3 extents
>>> 2.meg: 71 extents found, perfection would be 1 extent
>>> 32.meg: 46 extents found, perfection would be 1 extent
>>> 4.meg: 86 extents found, perfection would be 1 extent
>>> 64.meg: 96 extents found, perfection would be 1 extent
>>> 8.meg: 100 extents found, perfection would be 1 extent
>>>
>>
>> There again goes Hadron Quarks claim of "easy to allocate a
>> non-fragmented swap file". Which would be quite a bit larger than
>> 256meg
>
> What else would you expect from Tim "filesystem expert", "weekly stats
> analyst", "Funkenbush protege" Smith?
You look silly Roy trying to discredit them without offering an
explanation of the tests they ran.
So are they valid or not? (I have no idea).
If not, why don't you refute the data instead of attacking the messenger?
|
|
