On Sat, Oct 26, 2013 at 4:36 PM, Ray Van Dolson rayvd@bludgeon.org wrote:
On Thu, Oct 24, 2013 at 01:59:15PM -0700, John R Pierce wrote:
On 10/24/2013 1:41 PM, Lists wrote:
Was wondering if anybody here could weigh in with real-life experience? Performance/scalability?
I've only used ZFS on Solaris and FreeBSD. some general
observations...
- you need a LOT of ram for decent performance on large zpools. 1GB ram
above your basic system/application requirements per terabyte of zpool is not unreasonable.
- don't go overboard with snapshots. a few 100 are probably OK, but
1000s (*) will really drag down the performance of operations that enumerate file systems.
- NEVER let a zpool fill up above about 70% full, or the performance
really goes downhill.
Have run into this one (again -- with Nexenta) as well. It can be pretty dramatic. We tend to set quotas to ensure we don'get exceed 75% or so max, but....
We maybe getting a bit off topic here but on that subject we have noticed
a significant degrade in performance on systems running at 75-80 % of their pool capacity. I understand that the nature of COW will increase fragmentation. On large storages though 70% out of 100TB means that you have to always maintain 30TB free which is not a small number in terms of cost per TB.
- George Kontostanos ---