[svlug] More links about ssds

Marc MERLIN marc_news at merlins.org
Thu Jun 2 09:06:58 PDT 2016


On Thu, Jun 02, 2016 at 08:48:51AM -0700, Rick Moen wrote:
> Quoting Marc MERLIN (marc_news at merlins.org):
> 
> > I don't think any Sata SSD drive manufacturer guarantees any total
> > amount of data you can write to them, or did you find one?
> 
> I would never expect to.  Nor did I so suggest.
> 
> You might have missed my point, that the claim of 2000TB endurance is
> merely a reviewer's wild extrapolation from a small amount of SMART data
> gleaned during a brief run of test data on a single review unit -- and 
> so I put correspondingly little credence in that claim.
 
Ok, so at least we have this guy who tested the old generation:
http://techreport.com/review/27062/the-ssd-endurance-experiment-only-two-remain-after-1-5pb
http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

If we say about 1PB for the old 840 246GB, that's at least 8PB for the 2TB
and assuming the 850 is better than the 840, you ought to be able to count
on at least 10PB

> That is not _at all_ to denigrate the hardware in question, merely to
> point out that the data presented there are not compelling.

What's the write endurance of a recent 6TB spinning rust drive?
Considering how small the sectors are nowadays, it would be interesting to
see the write endurance of those drives too.

> > They offer estimates from wear levelling meters and people have reverse
> > engineered approximate numbers from them.
> 
> We're agreed that SMART data lies, right?  I mean, otherwise, we're 
> not having the same discussion.

Haha :)
Some do, some don't. Some data on some drives is actually pretty accurate
and useful.

Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/  



More information about the svlug mailing list