secure data erasing

Alison at wsrcc.com Alison at wsrcc.com
Wed Dec 7 21:09:56 PST 2005


alvin at Mail.Linux-Consulting.com (Alvin Oga) writes:
> but uv erasable eeprom "should be immune" to some magnetics but if
> the beam is strong enough, who knows what north pole will do under
> the extra strong magentic beams/pulses

I don't expect an arbitrarily high magnetic field to do much to flash,
which is an entirely electrostatic memory.  What a high magnetic field
can do is "upset" the read/write head.  A read/write head has
"biasing" magnets in it that set the operating point on the maximum
slope part of the response curve for highest possible sensitivity.  A
head also has stabilization magnets in it that keep the sensor in a
single magnetic domain state.  If either the biasing or stabilization
magnets in the head get screwed up, that would be the end.

> the itty-bitty section ofthe disk under the head at the time of the
> hammer might get destroyed, but the rest of data on the platter is
> still intact

As suggested above, destroying the heads (perhaps with tweezers) is a
lot easier than denting the platters with hammers.  Damaged heads
won't prevent someone from putting the platters in another system and
reading them, but then we're talking about a fairly motivated and
sophisticated snoop.

In my previous job in Washington DC, armored trucks would arrive from
an unknown agency with discs to be erased on a 20 tesla magnet.  (By
comparison the earth's field is about 0.3 G = 3x10-5 T.)  My lab was
unfortunately directly over this magnet, and naturally "they" would
never warn me in advance when they were coming.

-- 
Alison Chaiken    		"From:" address above is valid.
(650) 236-2231 [daytime]	http://www.wsrcc.com/alison/
Predators fail often; prey fail only once. -- Tom Evslin



More information about the Baylisa mailing list