secure erase on a SSD?

Yes, it will. How much depends on how often you do it. Nevertheless, it would probably be better to avoid Secure Erase. If you are using FileVault then your Home folder is encrypted, so using secure erase isn't really a necessity.

As I understand it secure erase is a waste of time on SSDs due to the way they write - basically the erase write pass is written to a different location to where the data was anyway, so don't bother.

Not sure about Encryption but I would have thought that wouldn't make a difference (others may know better).

SSD's have a finite amount of times they can be written to so it would depend on how much you use the "disk"

Regards,

Shawn

In addition to everything that Kappy had to say, there also is the problem that because of the method used to write data to the SSD to provide a longer life there are problems with removing the data in a secure manner. See the article in the link for additional information.

http://www.eweek.com/c/a/Data-Storage/SSDs-Harder-to-Seucrely-Purge-of-Data-than -HDDs-200129/

Allan

Encrypting an SSD for use with FileVault 2 has no significant effect on its lifespan. It writes to each cell on the drive once. The life expectancy of the drive is on the order of 100,000 write cycles per cell. The drive will become obsolete or fail for other reasons long before that limit even comes close to being reached.

That should be correct - AFAIK - FileVault (1 or 2) does not have many extra (if any at all) writes vs an 'open' file system ... and this would also (as others have said) remove the need to ensure a full secure erase of the entire drive ...

Also Disk Utily gives you the option to overwrite the data once or more times ... over writing the data once (if that is in fact what is doing) is fairly good ensurance that some snooper wont pull up your data with a simple program.

SRM also allows you to specify approximately what type of data overwrite occurs.. IIRC you have a simle (one random pass), medium (3 passes - one writes 0s, one writes 1s, and one is random) or the sandard 7 pass (DoD - spec for standard HDDs) ... and maybe even a 35 pass option that would probably take a week on most full HDDs.

The multipass is more important (but probably less so if it is simply peronsal finacial type data) since they can read weak singals coming off the heads (and maybe make something out of it ... even after the data is overwritten) ... though I think this is pretty expensive, time consuming and complex - since they are reading the analog signal off the drive heads... not the 1s & 0s the drive would have interpreted based on that signal.