Date: Wed, 23 Mar 2016 00:43:53 -0600 From: Warner Losh <imp@bsdimp.com> To: Jia-Shiun Li <jiashiun@gmail.com> Cc: "freebsd-arm@freebsd.org" <freebsd-arm@freebsd.org> Subject: Re: Effect of partitioning on wear-leveling Message-ID: <CANCZdfrJjCn%2Bo2C-QEYOMwD_8wiNM9qntGRzqVqpV1mh-HJQ1g@mail.gmail.com> In-Reply-To: <CAHNYxxMPHs4g=F7CqMEZTD23gfw%2Bb%2Bh1eTax=oUVFXW5UHQ03A@mail.gmail.com> References: <20160321175952.GA83908@www.zefox.net> <1458586884.68920.96.camel@freebsd.org> <20160321221153.GB83908@www.zefox.net> <1458600070.68920.107.camel@freebsd.org> <20160322032832.GC83908@www.zefox.net> <E985EBE6-E062-4C5E-8F85-ECB7BDE98DE8@bsdimp.com> <20160322062635.GD64087@server.rulingia.com> <CAHNYxxMPHs4g=F7CqMEZTD23gfw%2Bb%2Bh1eTax=oUVFXW5UHQ03A@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Mar 22, 2016 at 6:11 AM, Jia-Shiun Li <jiashiun@gmail.com> wrote: > On Tue, Mar 22, 2016 at 2:26 PM, Peter Jeremy <peter@rulingia.com> wrote: > > > On 2016-Mar-21 21:47:39 -0600, Warner Losh <imp@bsdimp.com> wrote: > > >So let=E2=80=99s do the math. 512MB cards tended to have write speeds = of maybe > > 6MB/s. > > >At 6MB/s, that=E2=80=99s about 518MB/day, or one drive write per day. = Most SD > > cards, > > > > I think you dropped some zeroes there. 6MB/s =3D=3D 518,400MB/day =3D= =3D > > 518GB/day. > > That's 1000 drive writes/day - which is non-trivial. > Yes, I must have. I think I must be misremembering the speed. Also, DD spee= d on the 512MB drives I just tried is closer to 3MB/s and writes through the file system are closer to 1MB for big writes and 500k for small writes... Still that's not enough of difference to make up for the while error. Maybe a factor of 10? > > > btw at the days of 512MB cards they are mostly made of SLC nand flash, > some were beginning to transition to MLC. They are different from TLC the= se > days in terms of endurance. That may explain it. SLC parts generally were good for 50k or 100k cycles, while MLC parts are good for 2k-5k. TTLC parts maybe 1k. The 1 drive write per da= y drives generally are at the low end of MLC or the high end of TLC. So that's another factor of 100 there maybe? Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfrJjCn%2Bo2C-QEYOMwD_8wiNM9qntGRzqVqpV1mh-HJQ1g>