[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: announce: readahead-1.4



On Fri, 2007-03-09 at 14:48 +0100, Benny Amorsen wrote:
> >>>>> "CL" == Callum Lerwick <seg haxxed com> writes:
> 
> CL> I don't see how they could patent defragging a disk. Lets not get
> CL> crazy here. ext3 does a decent job of not fragmenting files
> CL> unnecessarily, can we really gain much more than the current
> CL> readahead solution?
> 
> I think the operative word here is "unnecessarily". Desktop hard
> drives stay full once they fill up. Very few people clear off more
> than 20% once they've run out of disk space. Linux is better than most
> at avoiding fragmentation, but no (non-repacking) algorithm works well
> when only 5-20% disk space remains.
> 
> I really wish ext3 had a defrag utility which worked.

Yes, what it all comes down to, is we need a low level tool that can
safely move data blocks around on an ext3 filesystem, preferably while
mounted. On top of this we can build a whole filesystem defragger,
implement hot files, whatever.

Mainly what I'm trying to say is this flash crap Microsoft is hyping as
"cache" in Vista is a total farce. Flash is SLOW. I can't even think of
a good treacherous reason MS would push this bullshit on the industry.
So that Intel and Maxtor can sell more hardware once the flash wears
out? Does it enable Treacherous Computing in some way? I have yet to
stumble upon any conspiracy theories. Maybe I can be the one to start
some.

Now, journaling and hibernation, flash might have some use there. But
don't use it for cache. Go buy more RAM and defrag your disk, its
cheaper and orders of magnitude faster.

Attachment: signature.asc
Description: This is a digitally signed message part


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]