Date: Fri, 16 Sep 2011 23:47:00 +0200 From: Terje Elde <terje@elde.net> To: "freebsd@top-consulting.net" <freebsd@top-consulting.net> Cc: "freebsd-questions@freebsd.org" <freebsd-questions@freebsd.org> Subject: Re: FS of choice for max random iops ( Maildir ) Message-ID: <685DC833-382F-4AD6-B2F9-307645E7FB39@elde.net> In-Reply-To: <20110916101833.17485ybnq5srjbc4@mail.top-consulting.net> References: <20110916063153.200375qdq59crf8c@mail.top-consulting.net> <32990703-D068-4B0D-AF3A-C1E6EA0A4100@elde.net> <20110916101833.17485ybnq5srjbc4@mail.top-consulting.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 16. sep. 2011, at 16:18, freebsd@top-consulting.net wrote: > zpool create data da1 > zfs create data/maildomains > zfs set sync=3Ddisabled data/maildomains Just for the archives... sync=3Ddisabled won't disable disable the zil, it'l= l disable waiting for a disk-flush on fsync etc. With a battery backed contr= oller cache, those flushes should go to cache, and be pretty mich free. You e= nd up tossing away something for nothing.=20 You're getting about half the performance on a sequential write to the zfs, a= s you get with raw ufs. That makes perfect sense, doesn't it? Ufs writes raw, zfs writes to zil, then final restingplace forthe data. Acco= unt for the seeks between, and you're seeing what you should.=20 Move the zil if you don't want both those sets of writes on the same array, o= r do what Svein said, and get funk^w logical.=20 (a tad simplified, but I think the logic will hold. (yes, pun intended)) Terje=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?685DC833-382F-4AD6-B2F9-307645E7FB39>