From owner-freebsd-chat Mon Mar 2 16:45:32 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA29771 for freebsd-chat-outgoing; Mon, 2 Mar 1998 16:45:32 -0800 (PST) (envelope-from owner-freebsd-chat@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA29749 for ; Mon, 2 Mar 1998 16:45:12 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id TAA15081; Mon, 2 Mar 1998 19:44:39 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199803030044.TAA15081@dyson.iquest.net> Subject: Re: FreeBSD vs. Linux ? In-Reply-To: from Johann Visagie at "Mar 2, 98 08:34:00 pm" To: wjv@cityip.co.za (Johann Visagie) Date: Mon, 2 Mar 1998 19:44:39 -0500 (EST) Cc: chat@FreeBSD.ORG, marcus@vue.co.za X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Johann Visagie said: > John S. Dyson wrote: > > > > Much of the time, when people might say "Linux is faster than FreeBSD", > > it is likely a manifestation of the conservative filesystem metadata > > update policy of FreeBSD. This is (and don't let nay-sayers try to > > convince you otherwise) a carefully considered and wise policy on the > > part of the original BSD development group. > > [ rest of filesystem discussion deleted ] > > My apologies for posting a somewhat unrelated followup to this thread... > > Is there any specific reason why ext2fs would be inefficient at handling > large files (on the order of 1 meg and upwards to a few hundred megabytes)? > > We're doing some development which results in very large blobs of binary data > being bandied about, and I couldn't help but notice that FreeBSD machines > constantly seem to outperform higher specced Linux machines when doing > arbitrary file and directory manipulations (moving, removing, etc.) on these > large files, or even when merely accessing these files repeatedly. Sometimes > the difference is more than an order of magnitude. > I don't know, but my guess is it might be because of the indexing method used for file data. Also, our file/VM cache mgmt policy is not the usual one. That policy is designed to be high performance under actual high load conditions, while sacrificing some performance under low load conditions. The system does try to adapt automatically, but there are some things due to algorithmic complexity that will make FreeBSD appear to be slower in lightly loaded applications. Cache and memory management schemes sometimes do not perform in an intuitive fashion. Of course, under low load conditions, you don't always need quite as much performance :-). So, for server and heavily loaded workstation applications, I think that we made the right tradeoffs and generally maximized the spectrum of applications where FreeBSD is useful. -- John | Never try to teach a pig to sing, dyson@freebsd.org | it just makes you look stupid, jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message