Best practices for SSD

Chris Murphy lists at colorremedies.com
Sat Feb 8 00:54:31 UTC 2014


On Feb 7, 2014, at 5:44 PM, Chris Murphy <lists at colorremedies.com> wrote:

> 
> On Feb 7, 2014, at 5:11 PM, Michael Cronenworth <mike at cchtml.com> wrote:
> 
>> On 02/07/2014 03:00 AM, sam tygier wrote:
>>> 
>>> Ubuntu will be using a cronjob that calls fstrim, not the discard mount option.
>>> They found that the discard option cause a small performance hit.
>>> https://blueprints.launchpad.net/ubuntu/+spec/core-1311-ssd-trimming
>> 
>> If you read the link you posted you will find that their discard testing is inconclusive. They only tested 2 SSDs.
>> 
>> The whole discard vs. fstrim debate is entirely speculation. I haven't seen any fs maintainer speak up on either side either. The debate has been limited to blog posters and vocal end-users with their opinions.
> 
> Well it's an old topic for them, you have to go back 3+ years, which is maybe not what you expect when only in the last few months some loud mouths on /. and Engadget shriek ZOMG Linux just now is getting TRIM?! And how basic this is on Windows and OS X (despite OS X only enabling trim for Apple branded SSDs).
> 
> 
> Theodore T'so has spoken up on the ext4 list, as have others:
> http://www.spinics.net/lists/linux-ext4/msg13071.html
> 
> Dave Chinner has said some things about it on the XFS list, as have others:
> http://oss.sgi.com/archives/xfs/2013-07/msg00922.html
> 
> http://oss.sgi.com/archives/xfs/2011-05/msg00338.html
> 
> And there's a pile of discard/trim conversations on LKML, LVM, Btrfs, on and on.


And here is a recent thread, just under one year old….

https://lkml.org/lkml/2013/2/23/107

Anyway, the fs developers have been speaking about it quite a bit and for a long time.

Chris Murphy


More information about the users mailing list