From owner-freebsd-performance@FreeBSD.ORG Sat Jan 13 13:36:23 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id B06C216A412 for ; Sat, 13 Jan 2007 13:36:23 +0000 (UTC) (envelope-from patpro@patpro.net) Received: from smtp1-g19.free.fr (smtp1-g19.free.fr [212.27.42.27]) by mx1.freebsd.org (Postfix) with ESMTP id 7D97313C442 for ; Sat, 13 Jan 2007 13:36:23 +0000 (UTC) (envelope-from patpro@patpro.net) Received: from [192.168.0.2] (boleskine.patpro.net [82.235.12.223]) by smtp1-g19.free.fr (Postfix) with ESMTP id C4FB69B6A7 for ; Sat, 13 Jan 2007 14:36:22 +0100 (CET) Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: <45A81FA5.3090701@mawer.org> References: <7CA8AE1A-3925-404C-9F69-32AC4FFBB379@patpro.net> <45A81FA5.3090701@mawer.org> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Patrick Proniewski Date: Sat, 13 Jan 2007 14:36:21 +0100 To: freebsd-performance@freebsd.org X-Mailer: Apple Mail (2.752.2) Subject: Re: network perf : em driver ? [solved] X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 13 Jan 2007 13:36:23 -0000 Thank you all for your help. I've finally found the bottleneck: the onboard PowerMac ethernet port. After verification, changing network cables was not an option, I've used on this connection a cat. 7 network cable. I've connected the em0 port of the freebsd box on a gigabit PCI-X NIC I have in the Mac, and given a second try to the "dd | nc" client/ server bench: 103 MB/s. This is far better ! Same setup with apache 1.3 on the mac, wget on the Freebsd: about 40 MB/s (350 MB file) The PCI-X NIC has flow-control enabled: media: autoselect (1000baseT ) status: active I'll check later the influence of this parameter. Again, thank you all for your replies patpro