A litmus check request
bmettee at pchotshots.com
Tue Sep 9 16:08:05 UTC 2014
On 9/9/2014 11:20 AM, RW wrote:
> On Mon, 08 Sep 2014 15:37:36 -0600
> Dave Babb wrote:
>> I am about to do a lot of compiling. My copy of FreeBSD10.0 x64 P7
>> runs on a 1TB Samsung EVO. I used wblock's excellent articles to set
>> it all up so that trim was utilized.
>> I am asking that someone please review this attached fstab and
>> validate for me, that at no time will any compilation intermediate
>> files, or ".o" files will be written to the SSD. I am trying to
>> protect the SSD.
> If you haven't already done so I'd suggest you have a look at the
> output of smartctl -a <device> (from sysutils/smartmontools) and see
> if Samsung have added an attribute to track total writes into the NAND
> I recently installed an SSD, and found buildworld produced 6 GB
> of writes and buildkernel a further 6 GB. It's a substantial
> overhead, but not ruinous if you occasionally want to install world by
> rebooting into single-user mode. An alternative might be to build with
> MAKEOBJDIRPREFIX set to /tmp/obj and then copy /tmp/obj to /usr/obj.
> freebsd-questions at freebsd.org mailing list
> To unsubscribe, send any mail to "freebsd-questions-unsubscribe at freebsd.org"
The Tech Report has been an SSD Endurance Experiment since Aug 2013.
Three out of six starting drives have made it to the 1PB mark, the other
three all made it up to 600TB. All of the drives are consumer grade, not
Being careful is good, but depending on brand of drive, and several
other factors, the 200TB write life looks like be a bit of an
underestimate. As long as you try to reasonably minimize your writes to
the SSD, it doesn't look like you need to be too paranoid about it.
Links to updates on the experiment can be found at the bottom of the
last page (before the comments).
Hope this helps.
More information about the freebsd-questions