On Thu, Apr 06, 2000 at 06:56:47PM -0400, Lucky Green wrote:
> I am not aware of any high-end data recovery outfits that use software
> solutions. Everybody I know of in that space uses STM's. I believe it was
> Peter Gutmann who publicized the fact that you can buy STM workstations that
> ship with vacuum chucks for all popular platter sizes.
>
> --Lucky Green <[EMAIL PROTECTED]>
>
Have any of your contacts in this arena given you any sense of
how many layers of data their operational STM systems used day to day in
their recovery business (not some theoretical system they don't really
have up and usable) can actually recover off a typical disk platter ? Is
it 1, 2, 5, or 25 discrete layers ? And what kind of bit error rate in
the recovered date do they achieve with the STMs ? How automated is
the process ? Can they prepare a platter, pump down the chamber and read
out multiple layers of data almost as if reading a disk with the drive
electronics or is there a lot of human operator intervention and
twiddling required to set things up to retrieve a sector ? I assume the
actual interpretation of the STM scan output as encoded binary data is
completely automated and that they are not ever working from raster
images by hand using the human eye and brain as a kind of OCR (unlike IC
mask reverse engineering of a few years back) ?
Do they often recover overwritten information at all ? I would
imagine that most disk recovery work involves drives that went bad
leaving valuable data inaccessible via normal disk reading mechanisms
due to problems like corrupt servo tracks and damaged media surfaces
and heads rather than actual overwritten information. Sure there might
be cases of a sector or two that needs to be read in order to
correctly understand the rest of the data, but massive recovery
of gigabytes should be rare I would think...
--
Dave Emery N1PRE, [EMAIL PROTECTED] DIE Consulting, Weston, Mass.
PGP fingerprint = 2047/4D7B08D1 DE 6E E1 CC 1F 1D 96 E2 5D 27 BD B0 24 88 C3 18
