-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Benjamin, if you're using compression, forget about that. You need to synchronize the ashift value to the internal rowsize of you SSD, that's it. Make sure your SSD doesn't lie to you regarding writing blocks and their respective order. In that case you might even choose to set sync=disabled. Also, set atime=off and relatime=on. For faster snapshot transfers, you might like to set the checksum algo to SHA256. As always, put zfs.conf into /etc/modprobe.d with options spl spl_kmem_cache_slab_limit=16384 options zfs zfs_arc_max=8589934592 you might want to adjust the zfs_arc_max value to your liking. Don't set it to more than 1/3 of your RAM, just saying. I running above configuration in >30 servers atm in production, about 10 in test/dev environments, speed is awesome. No data loss so far, depite quite some brown/blackouts already. I'm using Ubuntu LTS (precise/trusty) with newest HWE, btw. hope that helps, Patric Benjamin Smith schrieb am 29.09.2015 um 22:08: > On Tuesday, September 29, 2015 10:35:28 AM John R Pierce wrote: >> On 9/29/2015 10:01 AM, Benjamin Smith wrote: >>> Does anybody here have any recommendations for using PostgreSQL 9.4 (latest) with ZFS? >> >> For databases, I've always used mirrored pools, not raidz*. > > >> put pgdata in its own zfs file system in your zpool. on that dedicated zfs, set the >> blocksize to 8k. > > Based on my reading here, that would be -o ashift=13 ? > HowDoesZFSonLinuxHandleAdvacedFormatDrives > > EG: 2^13 = 8192 > > > > -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) Comment: GnuPT 2.5.2 iEYEARECAAYFAlYLDtgACgkQfGgGu8y7ypBTywCfXvyWjmhAW+2AVl2ZVFBk45zy 190An1/OgNGHw7o48ZQiGQQbr2MTvqQ5 =yYUr -----END PGP SIGNATURE----- -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
