Date: Mon, 2 Dec 1996 23:34:17 -0500 From: "David S. Miller" <davem@jenolan.rutgers.edu> To: jgreco@brasil.moneng.mei.com Cc: jkh@time.cdrom.com, dyson@FreeBSD.org, dennis@etinc.com, kpneal@pobox.com, hackers@FreeBSD.org, torvalds@cs.helsinki.fi, lm@engr.sgi.com, iain@sbs.de, sparclinux@vger.rutgers.edu Subject: Re: TCP/IP bandwidth bragging Message-ID: <199612030434.XAA18481@jenolan.caipgeneral> In-Reply-To: <199612030319.VAA25727@brasil.moneng.mei.com> (message from Joe Greco on Mon, 2 Dec 1996 21:19:50 -0600 (CST))
next in thread | previous in thread | raw e-mail | index | archive | help
From: Joe Greco <jgreco@brasil.moneng.mei.com> Date: Mon, 2 Dec 1996 21:19:50 -0600 (CST) Now here is an intelligent posting. Let the jerk win, because clearly Linux is superior. We all run FreeBSD because it sucks rocks and because we like running something that is inferior. Benchmarks are always meaningful. Real world heavy duty applications mean nothing. I've said nothing that asserts the statements you are making. I've said nothing to the effect that "Using FreeBSD is not a good idea." or that doing so would get you less performance than Linux or any other system for that matter. As for real world heavy duty applications, I did give examples of where those are in use, but you can certainly feel free to ignore those statements as others have as well. Clearly I am not someone in the "real world" as my clients only invest six digit figures to implement my FreeBSD-based recommendations. You're one person, if you want to start talking market share then I'm game. But that is not much of a constructive thing to discuss. Marketing tally sheets are authoritative. So is the MIPS rating, a venerable benchmark. Let's all start using that! Regardless of whether they are authoritative or not, they do sell one machine over another, and often do translate into market share and installed base. Buying SPARC equipment is fiscally responsible. Yes I know, it drives me nuts how every major installation is composed of numerous Intel's running FreeBSD, very few Sparc's are to be found at all. But that's okay, because it's a free country. And he is free to believe whatever he wants, whether it is bullshit or fact. I have stated many facts, you have not refuted nor disproved any of them. The real shame is that this rivalry is antagonistic. As I have stated above I am not being very antagonistic. I have not denounced FreeBSD once as anything which is inferior, or that Linux is superior. At least not directly, and if people want to read between the lines and state that I meant for such things to come through in my arguments then that is fine, we can read into peoples wording all day. But we need to keep sight of the fact that FreeBSD and Linux are free software cousins - and I would rather see someone run _some_ free OS rather than a Gates Borg-osity. I do agree here, the Redmond crap is the real enemy indeed. Look, folks, benchmarks are benchmarks. They are not real world performance indicators. They are simply relative artificial performance evaluators, and as such can be influenced by a wide variety of factors, including OS tweaks. I agree one must be careful when analyzing the results of benchmarks, but I would not go so far as to call them completely artificial at all. That is a mistake and Mr. Dyson has made similar comments. I never make the mistake of taking a benchmark's results as an absolute comparison of apples and oranges. I have not stated that people _should_ do as such, but some people (and many of which decide what systems are to be run on what hardware platform) actually do. Hey, Linux may have some great benchmarks. I promise you that I can skew them in favor of FreeBSD. Hey, FreeBSD may have some great benchmarks. I promise you that I can skew them in favor of Linux. Hey, I can skew benchmarks to favor _SOLARIS_. Now there's a real performance lion! I've never skewed benchmarks, if you think I have then please support such claims. I'd be more than happy to be corrected. I run all of my benchmarks with both systems running on top of the same exact hardware configurations, sometimes the same exact machine using the same exact disk installed from scratch for both sides. How am I being impartial? Nobody will give a shit about any of this in a few years when we are looking at ten gigabit networking technologies. They will if people buy those pieces of hardware and nobody can fill the pipe. ---------------------------------------------//// Yow! 11.26 MB/s remote host TCP bandwidth & //// 199 usec remote TCP latency over 100Mb/s //// ethernet. Beat that! //// -----------------------------------------////__________ o David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612030434.XAA18481>