Date: Tue, 29 Dec 2009 11:35:03 +0100 From: "Ronald Klop" <ronald-freebsd8@klop.yi.org> To: "Thomas Burgess" <wonslung@gmail.com> Cc: freebsd-fs@freebsd.org Subject: Re: ZFS RaidZ2 with 24 drives? Message-ID: <op.u5o3opx08527sy@82-170-177-25.ip.telfort.nl> In-Reply-To: <deb820500912241646p7c8fd726n5c68895105e28907@mail.gmail.com> References: <568624531.20091215163420@pyro.de> <42952D86-6B4D-49A3-8E4F-7A1A53A954C2@spry.com> <957649379.20091216005253@pyro.de> <26F8D203-A923-47D3-9935-BE4BC6DA09B7@corp.spry.com> <1696529130.20091223212612@pyro.de> <op.u5gbvfgz8527sy@82-170-177-25.ip.telfort.nl> <deb820500912241646p7c8fd726n5c68895105e28907@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 25 Dec 2009 01:46:30 +0100, Thomas Burgess <wonslung@gmail.com> =20 wrote: > On Thu, Dec 24, 2009 at 11:53 AM, Ronald Klop > <ronald-freebsd8@klop.yi.org>wrote: > >> Isn't it write caching? >> My Solaris machine at work also flushes the data every 30 seconds. >> >> Ronald. >> >> I think you are right. ZFS does work in bursts...it's very different = =20 >> than > what most people expect. I know it was really weird to me when i firs= t =20 > saw > it. So increase your load and you will see a steady stream of bytes when all = =20 the bursts are clustered together. :-) BTW: it isn't that weird. UFS does the same, but you probably never build= =20 a UFS of 24 disks before. Ronald.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?op.u5o3opx08527sy>