Date: Tue, 22 Apr 2014 15:25:25 -0700 From: David Wolfskill <david@catwhisker.org> To: Louis Kowolowski <louisk@cryptomonkeys.org>, Julian Elischer <julian@freebsd.org> Cc: hackers@freebsd.org Subject: Re: Pointer to info on migrating from UFS2 -> ZFS? Message-ID: <20140422222525.GR1321@albert.catwhisker.org> In-Reply-To: <5355E9F9.5080401@freebsd.org> <63190425-672D-4A05-AAB0-B19A49EDB739@cryptomonkeys.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--uLzYCuFow5JXEQYy Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable I appreciate the responses, but I seem to have failed to communicate at least a couple of fairly important aspects of what I'm trying to do. So.... On Mon, Apr 21, 2014 at 06:40:05PM -0700, Louis Kowolowski wrote: > I?d probably suggest a couple things: > * VirtualBox (or equiv) for setting up test environments that are easy to= create and destroy. For all the beginning stuff I can think of, you should= be able to do just fine with a virtual environment. VMs with a half dozen = virtual disks that are 2G ea come in handy with playing with ZFS. I have existing hardware -- several instantiations of it, including a couple of test machines. I am trying to find out if the use of ZFS (vs. UFS2+SU) on the existing hardware will provide a performance advantage (and if so, how much, as switching from UFS2 to ZFS is going to be extremely painful). > ... On Tue, Apr 22, 2014 at 12:03:05PM +0800, Julian Elischer wrote: > ... > my experience so far is that ZFS is great if you want to do 'storage=20 > things' but that for > straight out speed you might want to stick with ufs+SU. > .... I appreciate that information, but I tend to be rather ... empirical ... about this kind of thing. Further, one of our colleagues was practically begging me to test ZFS, citing VFS lock contention as an observed bottleneck in the UFS2 case. Therefore, my intent is to try to set up a reasonably plausible ZFS environment -- to the extent that I am able -- using the existing hardware, and measure the workload in question (several times) in each environment. (I may change aspects of the configuration: for example, the UFS2 file system in question resides on a RAID 5 array of 10 spindles of 2TB each, using the "hardware RAID" capabilities of the controller. I understand that ZFS is better-suited to an environment in which the controller does as little as possible, so I would set that up and an N-spindle JBOD for ZFS. What to do about teh ARC is less clear to me -- as a specific case in point.) Peace, david --=20 David H. Wolfskill david@catwhisker.org Taliban: Evil cowards with guns afraid of truth from a 14-year old girl. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --uLzYCuFow5JXEQYy Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (FreeBSD) iQJ8BAEBCgBmBQJTVuxUXxSAAAAAAC4AKGlzc3Vlci1mcHJAbm90YXRpb25zLm9w ZW5wZ3AuZmlmdGhob3JzZW1hbi5uZXQ4RThEMDY4QTIxMjc1MDZFRDIzODYzRTc4 QTY3RjlDOERFRjQxOTNCAAoJEIpn+cje9Bk7V+4P/A5NhJiNT9Xir2gEsyjRY/i4 vRcOKXPICtpplwMFRS3tUXXBw99BAET+7GomthZQUXAGmK3vUbjfOsNnf3JySgvh e01MirdXa2mlHhvt1NrCK0X+U08kxJn7ws8ylU0rj/MkRbwGOyzdHuMqMHJaBFVQ qBuvu3fM2ViTlhQRrQbtqdofOcec2UhhZ+MlgAbqV0zNo9o4xnJQJr+bvjkhUadH uDRATHlG9U3d1GrPZ+vTXKBWd/why8tw6U8LDdJuOQYYbt1mW3izXvSZmLdzDDhT ZxV/5DPOTjBUmc8WYpCOaD2LWo6w4IWlNbEijcAIdusexGkFPFi3I0B3BjjHOuCv VKMi0KVSkqhb6FxEG48CTIIAQBGN+aiGwLbiXK3cNdcxvRFYDDQFdiqKRsGzjByN YzjpdFBDr+0QCvpe1wwSy+JWgU/GakaSmsOlEtARpCN3lrrsYpbZ/vBcBFDolQfP mTxdB1tze4Bmyr+VZnT50EwF+679Un/10OxA9ME8FVrYarwD2YI+zTVsdXpTnER9 xrBN4MEthpH28kGKqtlnrl3gU5cjebOwLbWMs8kgMHnQf90jfLFQ381QL5eBMvBt U8eeepmmqme4qrBQZUqi5uIMfiy39KqspG5uRA0hj84edtH/UH6iFDXpVkSmfiIt N8vfSgDK/F6h0V5tWhnS =jCHR -----END PGP SIGNATURE----- --uLzYCuFow5JXEQYy--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140422222525.GR1321>