Re: [luv-beginners] linux noob - install problems, installfests?

Hi Wen, Justin, On Mon, 8 Jun 2015 09:01:26 AM Wen Lin wrote:
Another key point: A Solid State Drive is worn down relatively quickly by write actions. So this site introduced steps/commands on how to minimise the action of writing to SSD by turning off the updating whenever a file is read (last accessed time) - using the parameter 'noatime'.
That's out of date - the kernel already defaults to "relatime" (since 2.6.30, released 6 years ago pretty much today), Red Hat describe it thus: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/htm... # Relatime maintains atime data, but not for each time that a file # is accessed. With this option enabled, atime data is written to # the disk only if the file has been modified since the atime data # was last updated (mtime), or if the file was last accessed more # than a certain length of time ago (by default, one day). So in other words for files that aren't being written you'll get an atime update at most once a day on reading it. commit 0a1c01c9477602ee8b44548a9405b2c1d587b5a2 Author: Matthew Garrett <mjg@redhat.com> Date: Thu Mar 26 17:53:14 2009 +0000 Make relatime default Change the default behaviour of the kernel to use relatime for all filesystems. This can be overridden with the "strictatime" mount option. Signed-off-by: Matthew Garrett <mjg@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Best of luck! Chris -- Chris Samuel : http://www.csamuel.org/ : Melbourne, VIC

Hi Chris, Thanks for your update. This is great! This means with the newer distros nowadays, we can basically treat SSD no different from conventional hard disks - when come to disk partitioning or formatting. I remember 2-3 years ago when I was installing Ubuntu onto my 2nd generation eeepc with SSD, I got instructions at the time to skip the swap space all together. Can I assume that with new distros on SSD - swap space is back in again? How about the "reserve 7% for overprovisioning" suggestion? Do you think it is still valid now? Anyway, the focus for Justin at the moment is to get his SSD to install Ubuntu 14.04 without error. Let us know how you go, Justin. :-) Wen On Mon, Jun 8, 2015 at 11:49 AM, Chris Samuel <chris@csamuel.org> wrote:
Hi Wen, Justin,
On Mon, 8 Jun 2015 09:01:26 AM Wen Lin wrote:
Another key point: A Solid State Drive is worn down relatively quickly by write actions. So this site introduced steps/commands on how to minimise the action of writing to SSD by turning off the updating whenever a file is read (last accessed time) - using the parameter 'noatime'.
That's out of date - the kernel already defaults to "relatime" (since 2.6.30, released 6 years ago pretty much today), Red Hat describe it thus:
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/htm...
# Relatime maintains atime data, but not for each time that a file # is accessed. With this option enabled, atime data is written to # the disk only if the file has been modified since the atime data # was last updated (mtime), or if the file was last accessed more # than a certain length of time ago (by default, one day).
So in other words for files that aren't being written you'll get an atime update at most once a day on reading it.
commit 0a1c01c9477602ee8b44548a9405b2c1d587b5a2 Author: Matthew Garrett <mjg@redhat.com> Date: Thu Mar 26 17:53:14 2009 +0000
Make relatime default
Change the default behaviour of the kernel to use relatime for all filesystems. This can be overridden with the "strictatime" mount option.
Signed-off-by: Matthew Garrett <mjg@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Best of luck! Chris -- Chris Samuel : http://www.csamuel.org/ : Melbourne, VIC

Wen Lin <vwenlin@gmail.com> writes:
How about the "reserve 7% for overprovisioning" suggestion? Do you think it is still valid now?
That sounds like bollocks to me. Or rather: specific to the make & model the author used. FTL over/underprovisioning is a knob you set before you ship to the retailer, and you can set it to anything you want. If you buy an SSD that presents (say) 16GB to the OS, and you tell the OS to only use the first 8GB, Filesystem Size Used Avail Use% Mounted on /dev/sda 8G 8G 1M 100% / ...then you'll get less wear than if you used all 16GB. But if you run fstrim --- which systemd does daily by default --- then with the same amount of actual files, you'll get about the same amount of wear, Filesystem Size Used Avail Use% Mounted on /dev/sda 16G 8G 8G 50% / ...because the OS will tell the FTL not to worry about old blocks. (Uh, YMMV. I'm a little bit too asleep to be confident about that assertion.) Also: general consensus is that you don't need to care about "wearing out" SSDs anymore.
participants (3)
-
Chris Samuel
-
trentbuck@gmail.com
-
Wen Lin