Mailinglist Archive: opensuse (2459 mails)

< Previous Next >
Re: [opensuse] scrathsing old harddisks
  • From: Sam Clemens <clemens.sam1@xxxxxxxxx>
  • Date: Thu, 27 Mar 2008 15:26:10 -0400
  • Message-id: <47EBF4D2.5050704@xxxxxxxxx>
Dave Howorth wrote:
Greg Freemyer wrote:
On Wed, Mar 26, 2008 at 11:33 PM, Sam Clemens <clemens.sam1@xxxxxxxxx> wrote:
You would be amazed at what can be accomplished with scanning
electron microscopes (due to the fact that the path of an
electron is effected by magnetic fields.). From what I
understand, due to hysteresis effects, a track starts out
at 'full width', but each time a magnetic field is reversed,
a "tail" is left on each side. Apparently, these residual
fields can be used to reconstruct data which was previously
overwritten.
I would love a true reference (from the last 15 years). I have spent
many hours looking into the question. The best I have seen is people
claiming they can recover a bit here and bit there from modern drives.
Not even any full bytes.

I have a NIST document that says labratory based recovery of data is
impossible for disk drives 20GB or larger if the have been overwritten
with a single pass of data. ie. Any data including all zeros.

It is certainly possible to read magnetic domain information from
surfaces using an AFM (atomic force microscope) in Magnetic Force Mode
(MFM). I've done that but not with a hard disk. If you google with those
keywords you can find images showing what can be seen - e.g. left over
data down the edge of tracks. But I don't know when those images were
made or what's possible or not with current hard disks.

I don't know about any equivalent SEM techniques.

My understanding is that any method which has fine enough resolution
to show the reversals of the magnetic domains is sufficient.
--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: opensuse+help@xxxxxxxxxxxx

< Previous Next >