[LDM #SJF-337060]: SSD for PQ

Clint,

The Wikipedia article <https://en.wikipedia.org/wiki/Trim_(computing)> has this 
to say about TRIM on a Linux system:

> Initial support for discard operations was added for FTL NAND flash
> devices in 2.6.28. Support for the ATA Trim command was added in 2.6.33.[25]

> Not all filesystems make use of Trim. Among the filesystems that can issue
> Trim requests automatically are Ext4,[26] Btrfs,[27] FAT, GFS2[28] and 
> XFS.[29]
> However, this is disabled by default due to performance concerns,[30] but can
> be enabled by setting the "discard" mount option. Ext3, NILFS2 and OCFS2 offer
> ioctls to perform offline trimming. The Trim specification calls for 
> supporting
> a list of trim ranges, but as of kernel 3.0 trim is only invoked with a single
> range that is slower.[31]

So, if you don't see the "discard" option in the mount(1) output, then it would 
appear that TRIM isn't enabled.

> No problem. As you noticed, it's doing well -- better than I expected.  Yes, 
> I did go
> with ext4 and the default mounting options.  I'm not exactly sure what those 
> are on our
> system and I'd like to find out if TRIM is enabled, but I'm not exactly sure 
> how to
> determine that, or even if it matters for the product queue (which is the 
> only thing on
> the SSD).  If you or Mike have some guidance on that, I'd certainly like to 
> hear it. I
> found a pretty good (I think) explanation/how-to at 
> http://blog.neutrino.es/2013/howto-
> properly-activate-trim-for-your-ssd-on-linux-fstrim-lvm-and-dmcrypt/.  I'm 
> not using
> dm-crypt (whatever that is) and there's only a single disk (other than the 
> SSD), so no
> LVM, either.  That leaves his recommendation of not using the discard option 
> in fstab
> and, instead, running fstrim periodically -- which he makes a good case for.
> 
> Clint

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: SJF-337060
Department: Support LDM
Priority: Normal
Status: Closed