a

  • Thread starter Deleted member 104402
  • Start date
There is no reason to think that the DOD standard is applicable to SSD, as they are an entirely different technology. I'm presuming that the DOD standard was developed by testing physical HDs.
 
It depends on how the testing is done.
If you are asking if SSDs are faster, then yes they are faster in terms of both read/write throughput.
But without any idea what the test procedure involves, how do you expect anyone to give you an answer unless they actually own an SSD and did that particular process on it?

You'll likely just have to buy one, do it yourself, and time it. If you saved time compared to the regular HD, then it's worth upgrading a whole lot of HDs to SSDs though presumably you'd still need to do that process with each new drive.
 
Data on HDDs is partially recoverable through up to 10 overwrites. I have no idea how the hell they do that, but that's what my evidence law prof said (former federal prosecutor).

I haven't seen or heard anything about what can be done with an erased SSD. I suspect it would be a lot harder to recover erased data off an SSD though. The main reason they can get data back off HDDs is "track wobble" -- the data isn't written in exactly the same place every time. SSDs have distinct physical bits, or distinct physical double bits in the case of MLC drives.

If you're just clearing the drive to sell it, I'd just use an appropriate utility to clear the drive. For Intel drives, you want to use the secure erase command. HDDErase (DOS) can do it, and hdparm (Linux) supposedly can, but I didn't have much luck finding a bootable CD/DVD that wouldn't error out. Some manufacturers offer erase utilities, and usually secure erase works as well unless you have a PITA laptop that freezes the ATA security commands on boot like mine does.
 
Data on HDDs is partially recoverable through up to 10 overwrites. I have no idea how the hell they do that, but that's what my evidence law prof said (former federal prosecutor).

I believe the general principle is that if a HD bit has been set at a certain level for long enough time, it's signal level will linger. Ex. suppose a bit is stored as either -1v or 1v. A file sits on the disk for a while, and then you erase it to zero (in this example). Maybe it erase to -0.2, and you can measure that the bit was probably a -1, because it lingered at the lower level. Now this is just speculation, and the track wobble thing is interesting. But that sounds more like something of magnetic tape than a HD to me.
 
I don't have much of anything to back that up. HDs do have wobble though. It's a mechanical device operating in very tight tolerances as quickly as possible. The whole drive is vibrating. (they make noise, therefore they vibrate) A new track won't land exactly on top of the last one except by random chance. The impression I got though was that the government could read data overwritten a fair number of times but wouldn't unless they really, really wanted to. In other words, it costs a lot and requires some skilled people and very expensive hardware. They don't just read the drive. The way I understood it, they take the platters out and use some sort of equipment to examine them. I have no idea what they actually do. For all I know they might be using something like an electron microscope. Equipment capable of detecting individual atoms exists, so I can't see why they couldn't get a reasonably detailed picture of a track of a magnetic disk. The tracks are a lot wider than a single atom.
 
Back
Top