Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 2 Mar 2012 08:28:50 -0800
From:      Freddie Cash <fjwcash@gmail.com>
To:        "Julian H. Stacey" <jhs@berklix.com>
Cc:        freebsd-fs@freebsd.org, Deb Goodkin <deb@freebsdfoundation.org>
Subject:   Re: [FreeBSD-Announce] Foundation Announces NAND File System for FreeBSD Project
Message-ID:  <CAOjFWZ4Xdk=fojhFLb7oufOaktAc9Phx748aW7=zSiu18=aVeA@mail.gmail.com>
In-Reply-To: <201203021530.q22FUjOU040478@fire.js.berklix.net>
References:  <4F4FC3E1.5040700@freebsdfoundation.org> <201203021530.q22FUjOU040478@fire.js.berklix.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Mar 2, 2012 at 7:30 AM, Julian H. Stacey <jhs@berklix.com> wrote:
> Hi freebsd-fs@freebsd.org
> cc Deb
>
> Deb Goodkin wrote to announce@ Thu, 01 Mar 2012 11:45:53 -0700 (19:45 CET=
):
>> The FreeBSD Foundation is pleased to announce that Semihalf, an
> ...
>
> I was interested to read more background, but no URLs,
> so I found http://www.semihalf.com/ & re.
> =C2=A0 =C2=A0 =C2=A0 =C2=A0NAND Flash File System
> found:
> =C2=A0http://en.wikipedia.org/wiki/Flash_memory#NAND_flash
> =C2=A0 =C2=A0 =C2=A0 =C2=A0"Manufacturers try to maximize the amount of u=
sable storage
> =C2=A0 =C2=A0 =C2=A0 =C2=A0by shrinking the size of the transistor below =
the size where
> =C2=A0 =C2=A0 =C2=A0 =C2=A0they can be made reliably, to the size where f=
urther
> =C2=A0 =C2=A0 =C2=A0 =C2=A0reductions would increase the number of faults=
 faster than
> =C2=A0 =C2=A0 =C2=A0 =C2=A0it would increase the total storage available.=
"
> =C2=A0 =C2=A0 =C2=A0 =C2=A0-
> =C2=A0 Really scarey if that is taken literaly, I jhs@ presume loose phra=
seology,
> =C2=A0 =C2=A0 =C2=A0 =C2=A0hopefully someone who knows more will rephrase=
 that :-).

A good primer on NAND Flash with lots of info on "the next hot thing",
triple-level cell (TLC) flash, which has all the flaws of MLC
multiplied many times over, but is a lot less expensive to use/make,
so expect it to hit consumer drives this year.  :(  Page 3 is
extremely enlightening about the issues, especially regarding the
increased CRC requirements in order to actually read a valid bit from
the cell.

--=20
Freddie Cash
fjwcash@gmail.com



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAOjFWZ4Xdk=fojhFLb7oufOaktAc9Phx748aW7=zSiu18=aVeA>