Date: Mon, 19 Sep 2011 11:49:45 -0700 (PDT) From: Jason Usher <jusher71@yahoo.com> To: Joshua Boyd <boydjd@jbip.net>, Julian Elischer <julian@freebsd.org> Cc: "freebsd-fs@freebsd.org" <freebsd-fs@freebsd.org> Subject: Re: ZFS obn FreeBSD hardware model for 48 or 96 sata3 paths... Message-ID: <1316458185.42258.YahooMailClassic@web121207.mail.ne1.yahoo.com>
next in thread | raw e-mail | index | archive | help
=0A--- On Fri, 9/16/11, Julian Elischer <julian@freebsd.org> wrote:=0A=0Awh= at is it you are trying to achieve?=A0 large storage, or high =0Atransactio= n rates? (or both?)=0A=0AI'm biased but I'd put a 160GB zil on a fusion-io = card=A0=A0=A0and dedicate =0A8G of the ram to it's useage.=0A=0Ait's remark= able what a 20uSec turnaround time on your metadata can do..=0A=0A---=0A=0A= I am optimizing for storage space - as much as possible for the budget.=0A= =0ASo we are going with SATA3 drives (the 3TB ones) and raidz3 (even though= it is a write-intensive application).=0A=0ABUT, in the context of those mi= serly constraints, I'm trying to figure out what mistakes to avoid and what= optimizations are worth making...=0A=0AThe fusion-io would be great, but a= nother $5k (or more) is not in the budget. I'm glad we can always add a ZI= L after the fact, and maybe we will do so with a cheaper (OCZ ?) pcie based= SSD...=0A=0A
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1316458185.42258.YahooMailClassic>