SATA problem

Jason M. Leonard fuzz at ldc.upenn.edu
Wed Mar 30 11:20:35 PST 2005


On Thu, 31 Mar 2005, Bill Hacker wrote:

Jason M. Leonard wrote:



On Thu, 31 Mar 2005, Bill Hacker wrote:

Jaime Andrés Ballesteros wrote:

Ok Bill, thanks for your answer. My SATA is a Seagate 7200.7 with
 120 GB. FreeBSD 5.3 doesn't recognize the geometry of my disk. FreeBSD 
force me to input another geometry and gives me 114 GB and i think this 
cause the problem.


Nope.  That is about all you will get.  Working with a pair of Western 
Digital 120 GB PATA the last 2 days, get about 114.4 GB with either 
FreeBSD or DFLY.  200 GB Maxtors, PATA or SATA give me about 190 GB or so.

Few HDD can (or should) give you 100% of the theorecticaly available area.


For marketing purposes 1G == 1000M, so with (120 x 1000)/1024 we're looking 
at a theoretically available area of 117G.  No one really makes 120G (117G) 
or 200G (195G) hard drives.

:Fu zz
Yes and 'not quite'.

Look at the slices and note that there is 'unused' (and unusable)
space showing.  Drives will reserve space for a pool of replacements for
bad blocks.  An OS will reserve a bit more.
Put the SCSI variant of any of the above on a sophisticated SCSI RAID
controller and use it unformated as a raw block device - the best way
for true hot-swap, wherein the MBR and partition table are stored in the
controller's NVRAM.  You can get a lot closer to the published capacity.
But such controllers are far and away more expensive than the miniscule
extra space one might gain.  That's not why you buy them.
1G == 1000M for marketing purposes is quite true; that is the published 
capacity.  The reserved space you are talking about is the difference 
betwen the 117G of theoretical space I was referring to and the 114G of 
available space referred to earlier by the original poster and yourself.

I wasn't trying to say you were wrong before; I was just addiing 
information :)

:Fuzz



More information about the Users mailing list