Date: Tue, 17 May 2016 15:29:20 +0300 From: Daniel Kalchev <daniel@digsys.bg> To: Alex Tutubalin <lexa@lexa.ru> Cc: freebsd-fs@freebsd.org Subject: Re: ZFS performance bottlenecks: CPU or RAM or anything else? Message-ID: <BD7DE274-04EB-4B19-988D-5A6FADC5B51A@digsys.bg> In-Reply-To: <16e474da-6b20-2e51-9981-3c262eaff350@lexa.ru> References: <8441f4c0-f8d1-f540-b928-7ae60998ba8e@lexa.ru> <f87ec54a-104e-e712-7793-86c37285fdaa@internetx.com> <16e474da-6b20-2e51-9981-3c262eaff350@lexa.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
> On 17.05.2016 =D0=B3., at 15:21, Alex Tutubalin <lexa@lexa.ru> wrote: >=20 > On 5/17/2016 3:11 PM, InterNetX - Juergen Gotteswinter wrote: >> Raidz is your Problem, go for Mirrors >=20 > Raidz2 will survive two (any) drives failure, while mirrored stripe = will not. >=20 Not true. You can have N-way mirror and it will survive N-1 drive = failures. > So, if it is possible to increase raidz2 performance by faster CPU or = RAM I'll go this route The limitations of RAIDZ performance do not come from CPU or RAM = limitations, but by the underlying hardware. RAIDZ is limited to the = performance of a single disk IOPS.=20 CPU/RAM these days are so much faster than spinning disks or SSDs. Daniel=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?BD7DE274-04EB-4B19-988D-5A6FADC5B51A>