Date: Thu, 22 Feb 2007 12:37:00 -0500 From: "Zaphod Beeblebrox" <zbeeble@gmail.com> To: freebsd-hackers@freebsd.org, peterjeremy@optushome.com.au, zbeeble@gmail.com Subject: Re: Fwd: Abyssmal dump cache efficiency Message-ID: <5f67a8c40702220937h21dc6963r77637ba369549e25@mail.gmail.com> In-Reply-To: <200702211113.l1LBDbQn006859@lurza.secnetix.de> References: <20070220182113.GC853@turion.vk2pj.dyndns.org> <200702211113.l1LBDbQn006859@lurza.secnetix.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2/21/07, Oliver Fromme <olli@lurza.secnetix.de> wrote: > > Peter Jeremy wrote: > > I've found that you do get a worthwhile improvement in dump|restore > > performance by introducing a large (10's of MB) fifo between them. > > This helps reduce synchronisation between dump and restore (so that > > dump can continue to read whilst restore is busy writing a batch of > > small files and vice versa). There's a suitable port but I can't > > recall the name because I wrote my own. > > There are several. The most popular ones are probably > misc/team and misc/buffer. I can certainly vouch for that , too. I generally use "team 1m 32" (total of 32meg of buffer). Team seems to not want to buffer more than 1m per process and I think 32 is the max # of processes.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5f67a8c40702220937h21dc6963r77637ba369549e25>