Hi Bob,

>> - “tank" is a raidz2 made up of spinning rust.
>> - I want to add a ZIL/Slog comprised of SSDs (INTEL 3700s)
>> - Being SSDs, "autotrim=on” should be set (right?)
> 
> It seems to me that the ZIL is not like a filesystem (where data is 
> continually allocated and freed) and it should be expected to appear to write 
> to the same regions over and over.  The FLASH should do wear-leveling as 
> erase areas start to wear out due to the number of writes.
> 
> Doing a trim of the ZIL might be counter-productive.  Doing a trim of normal 
> pool disks makes a lot more sense.

Interesting!

But the I now have the opposite problem!  ;)   As there is another zpool that 
consists a number of SSD-based vdevs (using IronWolf 110s).  As you say, this 
is a good case for "autotrim=on".  

This zpool also uses a ZIL/SLOG (using the same mirrored pair of Intel DC 3700 
SSDs*).  If what you say is true, than I’d want to have “autotrim=off” for the 
log vdev on this pool, right?, so I’m back to wanting to alter the “autotrim” 
value for specific vdevs.

* FWIW, the Intel DC 3700 SSDs are 400 GB, and hence too big for a single 
ZIL/SLOG.  Thus I partitioned each into six 60 GB partitions.  I’m using one 
pair of 60 GB partitions for the SSD-based pool's slog, one pair for the 
HDD-based pool’s slog, and the remaining four GB partitions are held in reserve 
for when the first two wear out.  Effectively, slog-specific spares.


Kent


------------------------------------------
illumos: illumos-discuss
Permalink: 
https://illumos.topicbox.com/groups/discuss/T6ef0a71646a80e63-M676ff95bcd1318cb2ab15b61
Delivery options: https://illumos.topicbox.com/groups/discuss/subscription

Reply via email to