On 13/02/2012 11:20, James Courtier-Dutton wrote:
On 13 February 2012 11:14, Anton Piatek wrote:
I didn't think you had direct access to sectors. I thought the controller
rotated data to spread writes across the the storage.
Yes, there is that, but if you write to the whole disk once, and th
As far as I am aware there remains a `wear-out' on Flash. I don't think
they've solved that.
I checked a few flash ICs that are likely to get used in SSDs and most
talk 100,000 cycles, have spare memory to replace failed bits and have
ECC in them.
That said, you know perfectly well, of course, t
On 13 Feb 2012, at 11:20, James Courtier-Dutton wrote:
> Yes, there is that, but if you write to the whole disk once, and then
> write to the whole disk a second time, you will be 100% sure that you
> have hit each actual flash sector at least twice, even if wear
> leveling is used.
Actually I be
I spent months worrying over what ssd's to buy, then i realised when
you compare the numbers to real world data...
it doesnt really matter, just buy the one you like the look of its
just as relevant!
--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman
On 13 February 2012 11:14, Anton Piatek wrote:
> I didn't think you had direct access to sectors. I thought the controller
> rotated data to spread writes across the the storage.
>
Yes, there is that, but if you write to the whole disk once, and then
write to the whole disk a second time, you wil
I didn't think you had direct access to sectors. I thought the controller
rotated data to spread writes across the the storage.
Anton
-
Anton Piatek
(sent from my phone, please excuse any typos)
email: an...@piatek.co.uk
blog/photos: http://www.strangeparty.com
pgp: [74B1FA37] (http:// www.strange
Hi,
I have looked at OCZ and Kingston datasheets, they mention things like
IOPS and read and write speeds, but there is no mention of "Max writes
per sector".
I thought that bit of info was useful because it helps you decide if
your application will wear the SSD out of not.
Is the reason why it is
