W dniu 2010-02-28 13:47, Latet pisze:
> Witam,
>
> Jak wiadomo (choć ciężko mi to pojąć) odzyskanie z dysku twardego danych
> nadpisanych innymi danymi jest możliwe (choć wymaga demontażu dysku i
> potraktowania talerza specjalnym urządzeniem - chyba to się nazywa
> mikroskop magnetyczny czy jakoś tak).
>
> A jak jest z nośnikami flash/ssd? Czy podobny zabieg jest możliwy, czy
> też jedno nadpisanie danych unicestwia jest całkowicie?

Nie jest tak jak piszesz. Wg. wikipedii, z dyskietek dało się jeszcze
coś odczytać po jednokrotnym zapisie, ale z dysków większych od 15GB
(duża gęstość zapisu), już jest to niemożliwe.

Wg. http://en.wikipedia.org/wiki/Data_recovery :

When data have been physically overwritten on a hard disk it is
generally assumed that the previous data are no longer possible to
recover. In 1996, Peter Gutmann, a respected computer scientist,
presented a paper that suggested overwritten data could be recovered
through the use of Scanning transmission electron microscopy.[8] In
2001, he presented another paper on a similar topic.[9] Substantial
criticism has followed, primarily dealing with the lack of any concrete
examples of significant amounts of overwritten data being
recovered.[10][11] To guard against this type of data recovery, he and
Colin Plumb designed the Gutmann method, which is used by several disk
scrubbing software packages.

Although Gutmann's theory may be correct, there's no practical evidence
that overwritten data can be recovered. Moreover, there are good reasons
to think that it cannot.[12]


http://en.wikipedia.org/wiki/Data_erasure :
According to the Center for Magnetic Recording Research, "Secure erase
does a single on-track erasure of the data on the disk drive. The U.S.
National Security Agency published an Information Assurance Approval of
single pass overwrite, after technical testing at CMRR showed that
multiple on-track overwrite passes gave no additional erasure."[7]
"Secure erase" is a utility built into modern ATA hard drives that
overwrites all data on a disk, including remapped (error) sectors.

Further analysis by Wright et al. seems to also indicate that one
overwrite is all that is generally required.[8]