From owner-freebsd-chat Thu Oct 8 16:09:00 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA14043 for freebsd-chat-outgoing; Thu, 8 Oct 1998 16:09:00 -0700 (PDT) (envelope-from owner-freebsd-chat@FreeBSD.ORG) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA13950 for ; Thu, 8 Oct 1998 16:08:34 -0700 (PDT) (envelope-from root@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with ESMTP id QAA00906; Thu, 8 Oct 1998 16:04:44 -0700 (PDT) Message-Id: <199810082304.QAA00906@implode.root.com> To: dag-erli@ifi.uio.no (Dag-Erling C. =?iso-8859-1?Q?Sm=F8rgrav?= ) cc: Eivind Eklund , Andre Oppermann , Reginald Perry , "'Jason C. Wells'" , freebsd-chat@FreeBSD.ORG Subject: Re: PC Magazine 10/20/1998 Article about FreeBSD In-reply-to: Your message of "08 Oct 1998 22:57:06 +0200." From: David Greenman Reply-To: dg@root.com Date: Thu, 08 Oct 1998 16:04:44 -0700 Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >Eivind Eklund writes: >> On Thu, Oct 08, 1998 at 09:19:59PM +0200, Andre Oppermann wrote: >> > Eivind Eklund wrote: >> > > On Wed, Oct 07, 1998 at 12:10:45PM -0700, Reginald Perry wrote: >> > > > So are you saying that you dont think that the network was saturated? >> > > Yes. 27Mb/s does not saturate a 100Mbit NIC, and 27Mb/s were >> > > (according to the author) what they got. >> > Did you say they used DELL boxes? I thought with DELL you'll get 3c509 >> > cards or something like that... >> You get either 3C905 or Intel EtherExpress 10/100+. They'd used Intel >> Etherexpresses. > >On a good day, my Etherexpress NICs (connected with a crossover Cat 5 >UTP cable) max out at about 23 Mbps, as measured by 'netstat -i 1' >while running 'rsh luna cat /dev/zero > /dev/null' from Niobe. Luna is >a P133 and Niobe is a P166. Running top on Luna shows 40% CPU usage >(most of it in cat, which spends a lot of time in sbwait) > >(running in 100baseTX full duplex of course) > >One is entitled to wonder where the bottleneck is - in the adapter, in >the bus or CPU, in the driver, or in some other part of the kernel. I have 2.1.x and 2.2.x machines here, Pentium (133-166), and all have Intel Pro/100B cards. I have no trouble getting >10MB/sec on these: [implode:dg] ttcp -p9 -n8192 -t core ttcp-t: buflen=8192, nbuf=8192, align=16384/+0, port=9 tcp -> core ttcp-t: socket ttcp-t: nopush ttcp-t: connect ttcp-t: 67108864 bytes in 6.49 real seconds = 10100.44 KB/sec +++ ttcp-t: 8192 I/O calls, msec/call = 0.81, calls/sec = 1262.55 ttcp-t: 0.0user 5.3sys 0:06real 83% 152i+341d 130maxrss 0+2pf 0+123csw 0.086u 5.377s 0:06.49 83.9% 153+341k 0+0io 0pf+0w [core:dg] ttcp -p9 -n8192 -t implode ttcp-t: buflen=8192, nbuf=8192, align=16384/+0, port=9 tcp -> implode ttcp-t: socket ttcp-t: nopush ttcp-t: connect ttcp-t: 67108864 bytes in 5.75 real seconds = 11400.66 KB/sec +++ ttcp-t: 8192 I/O calls, msec/call = 0.72, calls/sec = 1425.08 ttcp-t: 0.0user 3.2sys 0:05real 56% 143i+369d 76maxrss 0+2pf 13787+15csw -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message