From owner-freebsd-net@FreeBSD.ORG Thu Oct 12 08:38:55 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 E4EF716A403 for ; Thu, 12 Oct 2006 08:38:55 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from cs1.cs.huji.ac.il (cs1.cs.huji.ac.il [132.65.16.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6E3FA43D45 for ; Thu, 12 Oct 2006 08:38:55 +0000 (GMT) (envelope-from danny@cs.huji.ac.il) Received: from pampa.cs.huji.ac.il ([132.65.80.32]) by cs1.cs.huji.ac.il with esmtp id 1GXw5e-000O7H-ID; Thu, 12 Oct 2006 10:38:54 +0200 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.2 To: Mike Tancsa In-reply-to: References: Comments: In-reply-to Mike Tancsa message dated "Wed, 11 Oct 2006 13:28:58 -0400." Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Date: Thu, 12 Oct 2006 10:38:54 +0200 From: Danny Braniss Message-ID: Cc: freebsd-net@freebsd.org Subject: Re: em blues 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, 12 Oct 2006 08:38:56 -0000 > On Wed, 11 Oct 2006 16:06:17 +0200, in sentex.lists.freebsd.net you > wrote: >=20 >=20 >the box is a bit old (Intel Pentium III (933.07-MHz 686-class CPU) >=20 >dual cpu. >=20 > > >running iperf -c (receiving): >freebsd-4.10 0.0-10.0 sec 936 MBytes 785 Mbits/sec >freebsd-5.4 0.0-10.0 sec 413 MBytes 346 Mbits/sec >freebsd.6.1 0.0-10.0 sec 366 MBytes 307 Mbits/sec >freebsd-6.2 0.0-10.0 sec 344 MBytes 289 Mbits/sec > >btw, iperf -s (xmitting) is slightly better >freebsd-4.10 0.0-10.0 sec 664 MBytes 558 Mbits/sec >freebsd-5.4 0.0-10.0 sec 390 MBytes 327 Mbits/sec >freebsd-6.1 0.0-10.0 sec 495 MBytes 415 Mbits/sec >freebsd-6.2 0.0-10.0 sec 487 MBytes 408 Mbits/sec >so, it seems that as the release number increases, the em >throughput gets worse - or iperf is. > > Hi, > What is your setup for testing ? For me, with a couple of em > NICs back to back I get > iperf -c 1.1.1.2 > ------------------------------------------------------------ > Client connecting to 1.1.1.2, TCP port 5001 > TCP window size: 32.5 KByte (default) > ------------------------------------------------------------ > =5B 3=5D local 1.1.1.1 port 57584 connected with 1.1.1.2 port 5001 > =5B 3=5D 0.0-10.0 sec 1.06 GBytes 914 Mbits/sec >=20 > 6.2-PRERELEASE FreeBSD 6.2-PRERELEASE =230: Mon Oct 9 23:22:10 EDT 200= 6 >=20 > One is a Pentium(R) 4 CPU 3.00GHz and the other an AMD 3800 X2 >=20 > Going the other way is about the same (900Mb) >=20 > ---Mike no back to back, regular production infrastructure, Nortel Passport 8010 router as a backbone. one host, as mentioned, is a PIII, the other is an Intel(R) Xeon(TM) CPU 3.06GHz (3056.82-MHz 686-class CPU) running FreeBSD 6.2-PRERELEASE =239: Wed Oct 11 09:05:49 IST 2006 which gives nicer numbers, if for example the client is of the same hardw= are, and at the moment running 6.1-STABLE: 0.0-10.0 sec 1.08 GBytes 928 Mbits/sec short version: the point im trying to make, is that the same setup, where I only change the release, is going downhill - with this particular MB. the long version: Before I send this box to pasture, i decided to use it as a dns/dhcp/tfpt= server, and i also upgraded it to the latest/greatest version of freebsd,= before it becomes eol. as soon as it became production, i noticed that booting a class of some 60 ws was somewhat slower.=20 danny danny