From owner-freebsd-net@FreeBSD.ORG Thu Nov 16 22:35:10 2006 Return-Path: X-Original-To: freebsd-net@freebsd.org Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0436416A594 for ; Thu, 16 Nov 2006 22:35:10 +0000 (UTC) (envelope-from freebsd-net@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.FreeBSD.org (Postfix) with ESMTP id E7B3F43D80 for ; Thu, 16 Nov 2006 22:35:04 +0000 (GMT) (envelope-from freebsd-net@m.gmane.org) Received: from root by ciao.gmane.org with local (Exim 4.43) id 1Gkpp0-0007Zo-PW for freebsd-net@freebsd.org; Thu, 16 Nov 2006 23:35:02 +0100 Received: from mulderlab.f5.com ([205.229.151.151]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 16 Nov 2006 23:35:02 +0100 Received: from atkin901 by mulderlab.f5.com with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 16 Nov 2006 23:35:02 +0100 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-net@freebsd.org From: Mark Atkinson Followup-To: gmane.os.freebsd.current Date: Thu, 16 Nov 2006 14:30:47 -0800 Lines: 38 Message-ID: References: <455CB311.8040301@freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7Bit X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: mulderlab.f5.com User-Agent: KNode/0.10.4 Sender: news Cc: freebsd-current@freebsd.org Subject: Re: Automatic TCP send socker buffer sizing X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Nov 2006 22:35:10 -0000 Andre Oppermann wrote: > This is a patch adding automatic TCP send socket buffer sizing. Normally > the socket buffers are static (either derived from global defaults or set > with setsockopt) and do not adapt to real network conditions. Two things > happen: a) your socket buffers are too small and you can't reach the full > potential of the network between both hosts; b) your socket buffers are > too big and you waste a lot of kernel memory for data just sitting around. > > With automatic TCP send socket buffers we can start with a small buffer > and quickly grow it in parallel with the TCP congestion window to match > real network conditions. > > FreeBSD has a default 32K send socket buffer. This supports a maximal > transfer rate of only slightly more than 2Mbit/s on a 100ms RTT trans- > continental link. Or at 200ms just above 1Mbit/s. With TCP send buffer > auto scaling and the default values below it supports 20Mbit/s at 100ms > and 10Mbit/s at 200ms. That's an improvement of factor 10, or 1000%. > > New sysctl's are: > > net.inet.tcp.sndbuf_auto=1 (enabled) > net.inet.tcp.sndbuf_inc=8192 (8K, step size) > net.inet.tcp.sndbuf_max=262144 (256K, growth limit) > > The patch is available here: > > http://people.freebsd.org/~andre/tcp_auto_sndbuf-20061116.diff > > Any testers, especially with busy FTP servers, are very welcome. The patch at this address seems incomplete and only contains changes to tcp_output.c. Notably missing is the SB_AUTOSIZE bitfield definition. -- Mark Atkinson atkin901@yahoo.com (!wired)?(coffee++):(wired);