Date: Mon, 13 Jan 2003 17:16:30 +0100 From: Brad Knowles <brad.knowles@skynet.be> To: Pete Ehlke <pde@rfc822.net> Cc: freebsd-chat@FreeBSD.org Subject: Re: Need advice on PHP and MySQL books Message-ID: <a05200f09ba489479c2ca@[10.0.1.2]> In-Reply-To: <20030113155011.GA2719@rfc822.net> References: <20030110234309.R12065@2-234-22-23.pyvrag.nggov.pbz> <3E1FF12B.5390D978@mindspring.com> <20030111144619.X22424@2-234-22-23.pyvrag.nggov.pbz> <a05200f22ba474b755a2a@[12.27.220.113]> <3E21FD22.38CD81BB@mindspring.com> <a05200f02ba47b68abed1@[10.0.1.2]> <20030113141542.GC2260@rfc822.net> <a05200f07ba4881233aa2@[10.0.1.2]> <20030113155011.GA2719@rfc822.net>
next in thread | previous in thread | raw e-mail | index | archive | help
At 9:50 AM -0600 2003/01/13, Pete Ehlke wrote: >> I've benchmarked it, and it is *damn* bloody fast (see >> <http://www.shub-internet.org/brad/papers/dnscomparison/>). >> > Agreed. It simply blew me away when I tried it out here at home. It was > doing utterly unheard of numbers on old, low-end PC hardware. Yup. Using a private copy of the root zone and the .tv zone, I saw it doing 1400+ DNS queries per second, using queryperf (from BIND 9) on the same ultra-low end Compaq Armada 4131T laptop that I "inherited" from my wife when she upgraded (Pentium 133, 48MB RAM, 384MB virtual, 10GB IBM Travelstar 20GN, Asanté FriendlyNET AL1011 "Prism2" 802.11b WiFi NIC). I believe that my limitation here was how fast queryperf could generate the packets, not how fast NSD could respond. I tried doing some testing on my PowerBook G3 "Pismo" laptop running MacOS X 10.2.1 (400Mhz G3 processor, 1GB RAM, 2GB virtual, 48GB IBM Travelstar 40GH, Apple Airport 802.11b WiFi NIC), and I recall seeing NSD doing over 5000 DNS queries per second. Again, I believe that my limit was how fast queryperf could generate the questions, not how fast NSD could answer them. Simply unbelievable. > For those who might want to play with it, /usr/ports/net/nsd Excellent! How long has this been there? > Right. IIRC (and I may not; the coffee hasn't really kicked in yet...) > djb tried the same thing in early versions of his authoritative server, > and gave up on the idea. He's definitely done some whacked-out things. He's also claiming some equally unbelievable numbers. However, I have yet to see any proof of those claims. I'm a strong believer in the "show me" principle when it comes to claims of high levels of performance. In the case of NSD, I believe the numbers that I have personally seen, although I believe that the methods they used are not suitable for a general-purpose authoritative-only server, nor do they seem to be suitable for larger special-case situations such TLD nameservers and possibly root nameservers. -- Brad Knowles, <brad.knowles@skynet.be> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?a05200f09ba489479c2ca>