Text 46698, 103 rader
Skriven 2008-06-17 08:54:23 av Jeff Guerdat (1:123/789.0)
  Kommentar till text 46697 av Gerald Miller (1:342/512)
Ärende: MAX HDD size for Win2K?
===============================
Gerald Miller -> Jeff Guerdat wrote:

 JG>> The Microsoft-supplied generic IDE driver (Atapi.sys) may not be fully
 JG>> compatible with disks larger than 8 GB. This issue only affects
 JG>> IDE-based disks 8 GB and larger; however, Service Pack 4 includes an
 JG>> updated Atapi.sys which allows access to disks larger than 8 GB.
 JG>> Windows 2000 will support (bootable) partitions larger than 8 GB,
 JG>> providing the booting controller's BIOS supports EXT INT-13 BIOS
 JG>> calls.

 GM> So, the Windows 2000 install CD would very likely NOT have updated
drivers.
 GM> Could I perform the install on a small SATA HDD, go to the M$ web site to
 GM> get all the updates and fixes AND then use something like Norton Ghost or
 GM> Acronis True Image to clone the smaller drive onto a larger drive (and
 GM> still be able to access the unused space on the larger drive)?

The best thing to do would be to slipstream the latest service pack with the
original W2K install CD.  That will automatically give you the latest
atapi.sys, preventing other issues.

Using a SATA drive may introduce other issues, such as the lack of a SATA
driver for W2K - check Intel's web site for W2K drivers for your motherboard. 
Since these are newer boards, there may be no W2K drivers even available so you
could end up with either compatibility mode(s) (best case) or unknown/unusable
devices (worst case).  For compatibility mode, you may need to set things up in
the BIOS.  However, if the chipset doesn't have an INF for W2K, all bets are
off.

 GM> hard drives required for the three motherboards.  BTW, the three
DG965WHMKR
 GM> motherboards will do RAID, so I'm going to use four drives in each unit
 GM> with a RAID 10 configuration (where three drives are essentially one large
 GM> drive with the fourth drive for holding an image file.  I've been told
that
 GM> if one drive should fail, I can replace it and the image file will rebuild
 GM> the missing data onto the replacement drive.)  AND, no, I'm not planning
to
 GM> use this very large drive as my Boot drive.

Actually, using the RAID as the boot device helps get around the BIOS/atapi.sys
limitations, a la SCSI.  Since the driver is needed to boot in the first place,
there's no limitation - the BIOS emulation will get you to the device driver in
the very early stage of boot and then the driver takes over from there.

As for RAID 10, that's the same thing as RAID 1 plus RAID 0.  From
http://en.wikipedia.org/wiki/Redundant_array_of_independent_disks:

RAID 1+0: mirrored sets in a striped set (minimum four disks; even number of
disks) provides fault tolerance and improved performance but increases
complexity. The key difference from RAID 0+1 is that RAID 1+0 creates a striped
set from a series of mirrored drives. In a failed disk situation RAID 1+0
performs better because all the remaining disks continue to be used. The array
can sustain multiple drive losses so long as no mirror loses both its drives.

So, no, the effective drive is not 3 drives with parity (or, as you say, "the
fourth drive for holding an image file").  You're creating two mirrored sets of
two drives each (RAID 1) so your capacity for each set is only one drive (since
the other in the RAID 1 set is a mirror) x 2 (the stripe created by the RAID
0).  Using 100GB drives as an example, your 4 drives would be split into two
RAID 1 sets, each with 100GB capacity, and then striped with the RAID 0 for a
total capacity of 200GB.  The capacity is the same as using two 100GB drives in
RAID 0 only but you have fault tolerance that RAID 0 doesn't supply.  From
http://en.wikipedia.org/wiki/Nested_RAID_levels:

RAID 10 Capacity: (Size of Smallest Drive) * (Number of Drives) / 2

And:

All but one drive from each RAID 1 set could fail without damaging the data.
However, if the failed drive is not replaced, the single working hard drive in
the set then becomes a single point of failure for the entire array. If that
single hard drive then fails, all data stored in the entire array is lost. As
is the case with RAID 0+1, if a failed drive is not replaced in a RAID 10
configuration then a single uncorrectable media error occurring on the mirrored
hard drive would result in data loss. Some RAID 10 vendors address this problem
by supporting a "hot spare" drive, which automatically replaces and rebuilds a
failed drive in the array.

Given these increasing risks with RAID 10, many business and mission critical
enterprise environments are beginning to evaluate more fault tolerant RAID
setups that add underlying disk parity.

RAID 5, if available, may be a better solution since you gain capacity while
maintaining fault tolerance by using the 4th drive effectively for parity
information.  However, that depends on the RAID controller's capability to
supply RAID 5 as well as possible performance issues in creating the parity
information.  RAID 5 also only provides for a single drive failure, period,
while the RAID 10 scenario provides for limited ability to sustain multiple
failures.

Decisions, decisions.  It's only money...  ;)

-- 
Jeff Guerdat

The power of the executive to cast a man into prison without formulating
any charge known to the law, and particularly to deny him the judgement
of his peers, is in the highest degree odious, and the foundation of all
totalitarian government whether Nazi or Communist.     Winston Churchill

--- Thunderbird 2.0.0.14 (X11/20080501)
 * Origin: The Eastern Star - Fidonet Via Your Newsreader (1:123/789.0)