Re: Resources required to retrieved wiped disk

Paul Rubin wrote:
Andrew Swallow <am.swallow@xxxxxxxxxxxxxxx> writes:
Once you have recovered the random data you know its value. Then it
is time to discover the information it is hiding. This is an
iterative process that continues until the secret data has been
obtained.

Doh! Right, we're talking about a single overwrite pass. A double
overwrite would defeat this. However, even with a single overwrite I
think the first order prediction is not that useful. Drives use PRML
precisely because the amount of magnetization you get for any bit
varies from one place to another. It may also depend on what data was
there previously. I remember doing a calculation earlier in the
thread indicating that a 10 db decrease in total S/N ratio would make
recovery pretty hopeless. So we need this effect to get most of the
10 db back.

Check theory against experimental results.

Andrew Swallow
.

Relevant Pages

  • Re: Resources required to retrieved wiped disk
    ... is time to discover the information it is hiding. ... iterative process that continues until the secret data has been ... we're talking about a single overwrite pass. ...
    (sci.crypt)
  • Re: [opensuse] scrathsing old harddisks
    ... >> What is the best way to overwrite old harddisks with random garbage? ... > have not seen any docs that cover drives holding top secret data etc. ... > dollar recovery on and you think their is an ultra-secret government ...
    (SuSE)
  • Re: FileCopy overwrites the existing file
    ... This is why they still incinerate disks with secret data, ... the statement like " any good data recovery company can do it ... But if you want an empiric proof you can get it. ... Overwrite some data *once* take your hdd and request recovery of that data from a reputable company. ...
    (microsoft.public.win32.programmer.kernel)
Quantcast