From owner-freebsd-fs@FreeBSD.ORG Wed Jun 22 04:51:56 2011 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B791310657B5; Wed, 22 Jun 2011 04:51:56 +0000 (UTC) (envelope-from bsd@vink.pl) Received: from mail-bw0-f54.google.com (mail-bw0-f54.google.com [209.85.214.54]) by mx1.freebsd.org (Postfix) with ESMTP id 121668FC12; Wed, 22 Jun 2011 04:51:55 +0000 (UTC) Received: by bwz12 with SMTP id 12so671641bwz.13 for ; Tue, 21 Jun 2011 21:51:54 -0700 (PDT) Received: by 10.204.153.20 with SMTP id i20mr146657bkw.208.1308718314670; Tue, 21 Jun 2011 21:51:54 -0700 (PDT) Received: from mail-bw0-f54.google.com (mail-bw0-f54.google.com [209.85.214.54]) by mx.google.com with ESMTPS id v6sm161598bkf.11.2011.06.21.21.51.53 (version=SSLv3 cipher=OTHER); Tue, 21 Jun 2011 21:51:54 -0700 (PDT) Received: by bwz12 with SMTP id 12so671623bwz.13 for ; Tue, 21 Jun 2011 21:51:53 -0700 (PDT) MIME-Version: 1.0 Received: by 10.204.3.196 with SMTP id 4mr159493bko.188.1308718313469; Tue, 21 Jun 2011 21:51:53 -0700 (PDT) Received: by 10.204.152.212 with HTTP; Tue, 21 Jun 2011 21:51:53 -0700 (PDT) In-Reply-To: References: Date: Wed, 22 Jun 2011 06:51:53 +0200 Message-ID: From: Wiktor Niesiobedzki To: Artem Belevich Content-Type: text/plain; charset=UTF-8 Cc: freebsd-fs@freebsd.org Subject: Re: ZFS L2ARC hit ratio X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 22 Jun 2011 04:51:56 -0000 2011/6/22 Artem Belevich : > > L2ARC is filled with items evicted from ARC. The catch is that L2ARC > writes are intentionally throttled. When L2ARC is empty writes happen > at a higher rate, but it's still intentionally low so that > read-optimized cache device does not wear out too soon. The bottom > line is that not all the data spilled out of ARC ends up in L2ARC on > the first try. Re-run your experiment again and you would probably see > some improvement in L2ARC hit rates. I've run the experiment 3 times with no extent. Funny thing is: - in first run, I see a lot of write activity against cache device - in second run, I see no write activity against cache device, nor read activity So my guess is, that anyhow, ZFS cache layer knows, that this file is *there*, though it decides not to serve it from L2ARC... Cheers, Wiktor