Date: Thu, 28 Feb 2008 23:30:08 -0800 From: "Ted Mittelstaedt" <tedm@toybox.placo.com> To: "Kris Kennaway" <kris@freebsd.org> Cc: freebsd-questions@freebsd.org Subject: RE: FreeBSD bind performance in FreeBSD 7 Message-ID: <BMEDLGAENEKCJFGODFOCIEHACFAA.tedm@toybox.placo.com> In-Reply-To: <47C54201.5030706@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> -----Original Message----- > From: owner-freebsd-questions@freebsd.org > [mailto:owner-freebsd-questions@freebsd.org]On Behalf Of Kris Kennaway > Sent: Wednesday, February 27, 2008 2:57 AM > To: Ted Mittelstaedt > Cc: Oliver Herold; freebsd-questions@freebsd.org > Subject: Re: FreeBSD bind performance in FreeBSD 7 > > > >> * I am trying to understand what is different about the ISC > >> configuration but have not yet found the cause. > > > > It's called "Anti-FreeBSD bias". You won't find anything. > > This is false, but I didnt expect any better from you. > > ISC widely rely on FreeBSD internally, and contribute *lots* of > resources to the FreeBSD project including hosting one half of > ftp.freebsd.org and employing several FreeBSD developers. > So what? Microsoft has used FreeBSD in the past for it's DNS servers, and as far as I know still uses Linux or BSD for the nameservers for their download sites (or rather, the outsourcer they use doesen't use Windows for it's DNS) and they have never had anything good to say about FreeBSD - with the exception of the version 1 port of C# to it (which they dropped in verison 2) When Microsoft took over Hotmail, Hotmail was completely running on FreeBSD and several leaked internal documents showed many internal Microsoft people were highly impressed by FreeBSD when they got into it, but that didn't stop Microsoft from publically castigating FreeBSD in it's online "how we migrated Hotmail to the (superior) Windows platform" whitepapers. Apple's dependence on FreeBSD is legendary - yet Steve Jobs has several times at MacWorld referred to Darwin as based on Linux and that it's Linux-like, and similar to Linux, all of which are baldfaced lies. (At least, according to the Apple website which credits FreeBSD here: http://developer.apple.com/opensource/index.html) The point here is there are MANY organizations that publically beat the Linux bandwagon drum yet privately don't use Linux as much as they use FreeBSD "internally" This "study" of theirs on http://new.isc.org/proj/dnsperf/ is the proof of the pudding. I also noticed according to the testbed they are using HP Proliant DL140 G3 servers - those servers use El-crappy Broadcom 5722 ethernet chips on their motherboard, and the FreeBSD driver for these chips is iffy - FreeBSD 6.1 in fact paniced when using this chip family, as I documented in a PR for an HP Proliant server. And, HP supports and supplies the RedHat Linux driver for this chipset for this server, and there's no question that Gentoo uses the same driver. I can hardly think of a more unfair testbed that is more tilted towards Linux than these servers. But that's OK you continue rooting around in the FreeBSD 7 kernel all you want, don't bother actually looking at the network hardware, we all know it doesen't matter.....NOT! Consider also that ISC is 501(c)(3) The money they are spending on employing FreeBSD developers and hosting ftp.freebsd.org isn't theirs. It's donated to them specifically to be used for these purposes. > >> e.g. NSD > >> (ports/dns/nsd) is a much faster and more scalable DNS server than BIND > >> (because it is better optimized for the smaller set of features it > >> supports). > >> > > > > When you make remarks like that it's no wonder ISC is in the business > > of slamming FreeBSD. People used to make the same claims about djbdns > > but I noticed over the last few years they don't seem to be doing > > that anymore. > > What, you mean "factual statements"? NSD *is* faster, it *is* more > scalable, it *does* support fewer features than BIND, and it *is* more > optimized for those features (e.g. it tries to precompute DNS responses, > which it can do because it doesn't support dynamic updates, etc). The > ISC devels acknowledge this. BIND has architectural constraints from > being a more complete DNS server solution. > You could have just as easily said "a more feature-lacking, stripped-down nameserver like nsd is faster." That's factual too. Your even doing it now; "architectural constraints?" Look at the language the ISC uses to describe it's server: http://www.isc.org/index.pl?/about/press/?pr=2007032700 I see "fastest version yet" not "slower than other nameservers" Granted, neither group is making money on the nameserver software, it's not like money is at stake here. But, pride is. Ted
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?BMEDLGAENEKCJFGODFOCIEHACFAA.tedm>