On 11/30/22 20:51, Ron wrote:
ZFS will yes be slower than a raw disk (but that’s not an option for Pg anyway), and may or may not be faster than a different filesystem on a HW RAID volume or storage array volume. It absolutely takes more care/clue/tuning to get Pg write performance on ZFS, and ZFS does duplicate some of Pg’s resiliency so there is duplicate work going on.
Ron, if this raw device reference is about ASM, Oracle has a file system on top of ASM, called ACFS, and I have been able to run PostgreSQL on top of ACFS. The reason to do that is redundancy. ASM/ACFS with PostgreSQL behaves similarly to Veritas Cluster, when one Postgres cluster goes down, the other one is started. And you don't have to pay for it, unless you start using storage snapshots. That ACFS feature requires a commercial license.
--
Mladen Gogala
Database Consultant
Tel: (347) 321-1217
https://dbwhisperer.wordpress.com