From owner-freebsd-hackers Thu Jan 18 05:00:20 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id FAA06356 for hackers-outgoing; Thu, 18 Jan 1996 05:00:20 -0800 (PST) Received: from mailx.best.com (mailx.best.com [204.156.128.56]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id FAA06350 for ; Thu, 18 Jan 1996 05:00:18 -0800 (PST) Received: from geli.clusternet (rcarter.vip.best.com [204.156.137.2]) by mailx.best.com (950911.SGI.8.6.12.PATCH825/8.6.5) with ESMTP id NAA24256; Thu, 18 Jan 1996 13:02:19 GMT Received: from localhost (localhost [127.0.0.1]) by geli.clusternet (8.6.12/8.6.9) with SMTP id EAA07365; Thu, 18 Jan 1996 04:59:43 -0800 Message-Id: <199601181259.EAA07365@geli.clusternet> X-Authentication-Warning: geli.clusternet: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.4 10/10/95 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: freebsd-hackers@freebsd.org (FreeBSD hackers) Subject: Re: BSDvs Lxxxxx Flame.. In-reply-to: Your message of "Thu, 18 Jan 1996 10:16:26 +0100." <199601180916.KAA25329@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 18 Jan 1996 04:59:43 -0800 From: "Russell L. Carter" Sender: owner-hackers@freebsd.org Precedence: bulk } As Marty Leisner wrote: } > } > } > Can someone provide hard information about nfs serving capability } > on identical hardware between linux and freebsd? } > } > I'm very disappointed at the nfs performance on linux...I've used } } No ``hard information'', but most linux guys admit that their NFS } server is about the worst piece of the system. } } I've seen 800 KB/s being piped out of a FreeBSD 1.1.5.1 machine, going } to a single SGI Indy. } Asynchronous, of course! :-) I have a page of results for FreeBSD networking, sitting at http://www.geli.com/data/net.perf.html This is all 100BASE-TX results. I made a careful study of Linux, NetBSD, and FreeBSD 10Mb NFS Server performance for Sandia National Lab a year ago, and Linux was 10x slower. That was then, this is now, the first thing that happens when Linux server performance is mentioned is "well, this is fixed in the next kernel". But it isn't, the server is a user space implementation, and performance won't get better until they bring it into the kernel. Rather than have a flame war, the best antidote to the disinformation (dissembling?) proclivities of ->OTHER FINE PROJECTS<- is good hard data. If anyone has any networking (or any FreeBSD performance data) that they would like mentioned or listed (with proper attribs, of course) in my performance pages, I would be happy to due so. A good example how this sometimes works is disk performance. About 9 months ago I started pointing out on the linux lists that bonnie with a file size quite a bit larger than main memory size is a good indicator of actual disk performance, and slowly that discussion has gotten more realistic. (fewer "Zowwy! I get 15 MB/s out of my EIDE drive on a 486!) The same needs to be done for the networking side. I'm quite partial to netperf, myself. } -- } cheers, J"org } } joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE } Never trust an operating system you don't have sources for. ;-) } Regards, Russell