** This is the quasi-official and semi-temporary T13 email list server. **
Hi Jim
Thanks for your reply, I don't want to "spam" this forum
with this subject but I tell you my remark :
I completely understand that drive manufacturers
needs to hide information, and that the tendency is to
provide easy to use interface to read and write sectors.
I can suppose there are also some patents in the way
each one decides to allocate LBAs on the platters.
I just speak about the specific point of view of the
defragmentation tool : I have the feeling of doing somewhere
an incomplete work when defragmenting, on the LBA order only.
Gilles
----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, May 21, 2001 8:48 PM
Subject: [temp t13] Re: ARRRGH: Drive performance
** This is the quasi-official and semi-temporary T13 email list server. **
Gilles,
The fundamental issue is layering and information hiding. Exposing what
HDDs do internally to the outside world allows more software/APIs at the
system level to extend down into the drive. This in turn has a tendency to
retard the independent development of both over time.
This is important since for many things the HDD manufacturers are far better
suited (mainly due to size, expertise, and motivation) to develop the
technology than are system software folks. Take defect management. All
OSes have had some form of it since the beginning (usually they put "bad"
sectors in file that only the system can see). However, the implementations
in HDDs are far more robust. In addition to receiving a lot more testing,
the HDD manufacturer has a far greater understanding of the underlying cause
of media defects (for instance, can often distinguish it from transient
errors).
To take a more extreme example, HDDs use to lack formatting (that was
supplied by separate devices called controllers), including a lack of ECC.
Could anyone make the case that another company is better suited to
determining the ECC needs than the company which designed the drive? ECC
can change every year as technology matures. If system level software
depended upon knowing exactly how ECC works, then it would have to be
updated at least every year.
This is evident in other areas of the computer industry, where a poor
decomposition of tasks have resulted in the freezing of technology due to
the inability of companies to innovate freely and independently. There is
never a perfect answer to this, but creating dependencies in areas where
there are few functional complaints today is certainly not a good answer.
Jim
-----Original Message-----
From: Gilles Mollié [mailto:[EMAIL PROTECTED]]
Sent: Sunday, May 20, 2001 11:38 PM
To: [EMAIL PROTECTED]
Subject: [temp t13] Re: ARRRGH: Drive performance
** This is the quasi-official and semi-temporary T13 email list server. **
Hi Hale.
Yes of course I know ATA should drive something different that
a disc, without cylinders zones and platters but ...
For now, I think we can all agree a good defragmentation works
on ascending LBA (or descending). A good result is when a file
is recorded in huge contiguous LBAs.
Well ... even if the File System should only manage something of
quite high level as LBA, don't you think the efficiency of the drive
itself should be really improved if having another entry point for all
those optimizing HDD performance software ?
I find strange that such software cannot know that a defective sector
has been reallocated several tracks further.
Immagine also the work of people having to test and compare several
HDD. All the tests will perhaps give a bad result on one drive just
because of one sector reallocation ...
Solution could be a little ATA-7 (not for now of course) extension ... ?
Thanks for reading.
Gilles Mollié
----- Original Message -----
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, May 18, 2001 7:36 PM
Subject: [temp t13] ARRRGH: Drive performance
> ** This is the quasi-official and semi-temporary T13 email list server.
**
>
> I really do hate to discuss disk drive performance, however, ...
>
> And I will apologize now for being so, how shall I say this,
> rude? But these sort of statements really bug me!
>
> >[EMAIL PROTECTED] said: "In general you are correct,
> >but then there are exceptions. Some people like to fully
> >characterise HDD's by measuring head & cylinder skews, transfer
> >rates by zone, HDD command overhead, seek time versus seek
> >distance etc etc. In these cases such an interface provides the
> >accuracy required."
>
> Even if you think you can measure head and cylinder (oh, don't
> forget zone) skews from the interface using standard commands (I
> hate to tell you this but this is big waste of time), how do you
> know the data you get from one drive will look anything like the
> data you get from another drive of the same brand and model and
> having a serial number one greater? YOU DON"T! You have no idea
> how the factory defects are distributed, no idea how the sectors
> are arranged (and this is not a simple as you think it is). And
> I wonder how many people realize that lower capacity drives are
> just that because the drive would have been a full capacity drive
> except one of the heads was defective? Please, when you make
> statements about "characterize" a drive it tells me you really
> don't understand disk drives or what you are measuring. How do
> you even know if a drive is "seeking" between two LBA addresses?
> You might be surprised to find that two LBA addresses that are
> 1000's or 100000's or even 1000000's apart map to physical
> sectors that are very close on the media. What you think should
> be a big seek might not be a seek at all.
>
> ...
--
If you have any questions or wish to unsubscribe send a
message to Hale Landis, [EMAIL PROTECTED] To post to
this list server send your message to [EMAIL PROTECTED]
For questions concerning Thistle Grove Industries or TGI's
list services please send email to [EMAIL PROTECTED]
--
If you have any questions or wish to unsubscribe send a
message to Hale Landis, [EMAIL PROTECTED] To post to
this list server send your message to [EMAIL PROTECTED]
For questions concerning Thistle Grove Industries or TGI's
list services please send email to [EMAIL PROTECTED]
--
If you have any questions or wish to unsubscribe send a
message to Hale Landis, [EMAIL PROTECTED] To post to
this list server send your message to [EMAIL PROTECTED]
For questions concerning Thistle Grove Industries or TGI's
list services please send email to [EMAIL PROTECTED]
