Date: Sun, 13 Jul 2014 22:33:52 -0300 From: Joseph Mingrone <jrm@ftfl.ca> To: freebsd-questions@freebsd.org Subject: Re: deciding UFS vs ZFS Message-ID: <86sim4yb9b.fsf@gly.ftfl.ca> References: <20140713190308.GA9678@bewilderbeast.blackhelicopters.org> <20140714071443.42f615c5@X220.alogt.com> <53C326EE.1030405@my.hennepintech.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
Andrew Berg <aberg010@my.hennepintech.edu> writes: > On 2014.07.13 18:14, Erich Dollansky wrote: >> use UFS as long as you are working with a single disk and ZFS the >> moment you have more than one disk. > Checksumming and the COW features make ZFS quite attractive for single-device > pools as well. Off the top of my head, there is also on-the-fly compression, snapshots and boot environments. The way pools/datasets don't require you to decide how much space is allocated for a dataset at creation time is nice. I am happy I chose ZFS for my laptop with 8 GiB of ram and a single SSD. It works very well, just like our compute node with 256 GiB of ram, 48 cores connected to a 40-disk chassis. The way I think about it is more this way: "When can't I run ZFS?". The answer is when system resources simply don't allow it. Joseph
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86sim4yb9b.fsf>