Groups | Search | Server Info | Login | Register
Groups > alt.comp.hardware > #20731
| From | Paul <nospam@needed.invalid> |
|---|---|
| Newsgroups | alt.comp.hardware |
| Subject | Re: How Do SSDs Wear Out? |
| Date | 2025-02-18 03:03 -0500 |
| Organization | A noiseless patient Spider |
| Message-ID | <vp1etd$1k2i0$1@dont-email.me> (permalink) |
| References | <XnsB285DDB4AC76FBorisinvalidinvalid@135.181.20.170> <vomtr3$3cnqe$1@dont-email.me> <XnsB289B41E2295ABorisinvalidinvalid@135.181.20.170> |
On Mon, 2/17/2025 8:42 PM, Boris wrote:
>
> Thanks much for the education. I've read it over many times, and it's
> taken me to all sorts of articles, starting with those on floating gate
> transistors.
>
> https://tinyurl.com/zcj5j4d7
>
> I have a question:
>
> Does each cell have only one bit ("1" or "0") of changeable information?
>
Each cell is analog. It has a voltage on it. If a cell
is 5/8th full, that is equivalent to 101 binary. I can
store three bits of information, if I encode as eight
voltage levels within a single cell.
The claimed limit for such a technique at the moment,
is five bits per cell, or 32 voltage levels for the
analog voltage store of a cell. This has not been
delivered yet.
The error correcting code, is powerful. And the size of
the syndrome (50 bytes per 512 byte sector) is a heavy
tax for the design to use. When the five-bit-storing cell
comes along, how long will the syndrome need to be ? It
will be a significant portion of a sector size. This
is what prevents this game from "going to infinity",
is the need for more and more powerful error correction.
The error correction is not a hard macro (logic gates).
This is why the SSD has ARM CPU cores and error correction
is done in firmware. Most of the cores work on error correction.
An SSD can have a three or four core CPU in it.
The noise margins on cell storage are going to be so poor
at some point, when you write a sector, reading it back
will already have an error in it. You won't need to wait
for the cells to become "spongy", they will already be making
errors.
SLC \___ good storage \____ Not a lot of these chips
MLC / and lots of writes / are being manufactured
TLC ... acceptable, much of mainstream SSD uses this
QLC ... hanging on
PLC ... practicality, to be determined (not shipping now)
Paul
Back to alt.comp.hardware | Previous | Next — Previous in thread | Find similar
How Do SSDs Wear Out? Boris <Boris@invalid.invalid> - 2025-02-14 05:47 +0000
Re: How Do SSDs Wear Out? Paul <nospam@needed.invalid> - 2025-02-14 03:11 -0500
Re: How Do SSDs Wear Out? "Carlos E.R." <robin_listas@es.invalid> - 2025-02-17 21:46 +0100
Re: How Do SSDs Wear Out? Paul <nospam@needed.invalid> - 2025-02-18 00:28 -0500
Re: How Do SSDs Wear Out? "Carlos E.R." <robin_listas@es.invalid> - 2025-02-18 14:26 +0100
Re: How Do SSDs Wear Out? Boris <Boris@invalid.invalid> - 2025-02-18 01:42 +0000
Re: How Do SSDs Wear Out? Paul <nospam@needed.invalid> - 2025-02-18 03:03 -0500
csiph-web