Date: Thu, 15 Apr 1999 12:29:31 +0200 (MET DST) From: Emmanuel DELOGET <pixel@DotCom.FR> To: syssgm@detir.qld.gov.au (Stephen McKay) Cc: hackers@FreeBSD.ORG (FreeBSD Hackers Mail List) Subject: Re: NT4 server 2.5 times faster than Linux Message-ID: <199904151029.KAA14859@excalibur.oceanis.net> In-Reply-To: <199904150921.TAA28780@nymph.detir.qld.gov.au> from Stephen McKay at "Apr 15, 1999 7:21:26 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
As the well known Stephen McKay said... ->On Wednesday, 14th April 1999, Luoqi Chen wrote: -> ->>Saw this on yesterday's slashdot news: ->> http://www.mindcraft.com/whitepapers/nts4rhlinux.html ->>I wonder how well FreeBSD would perform. -> ->I wonder also. The load is much higher than similar NT vs Linux tests ->that I've seen. And the hardware is all very, very new. -> ->People shouldn't be so quick to dump on Mindcraft, even if they were ->bought by Microsoft. The facts of the report should be digested and ->argued rationally. Then, if we find that Mindcraft have deliberately ->distorted things, we can dump on them. :-) -> ->So far I've spotted only 2 relevant facts: Note a third on the apache configuration : 3) Set OPTIM = "-04 -m486" before compiling Will you choose 486 optim on a 4xPII Xeon ? I would not, since this will slow down the whole program... But the goal was not to have it to run fast... On a similar point of view, It could be nice to have the complete configuration of the linux box (maybe they have choosen 486 cpus too :) And More in IIS : 4)Server set to maximize throughput for applications when doing WebBench tests No comment about this... In fact, we don't care that linux/apache is slower than nt/iss, since linux/apache is able to run without any reboot for months, and nt/iss is not (I experienced 5/10 reboot a week during 3 months in a small french ISP). -> ->1) "Drive D/Data: 8 x 4 GB Seagate Barracuda, Model ST34573WC, 7,200 RPM; -> two partitions - one data partition for each OS" -> ->The two operating systems shared the drive array, so one of them got the ->good bit, and one got the bad bit. This is flawed testing. -> ->2) "The Linux kernel limited itself to use only 960 MB of RAM" -> ->The box had 4GB of RAM, but Linux got to use less than 1GB. Poor Linux. ->This was such a fair test! :-( Do we recall a previous test where our ->favourite OS used only a portion of the total RAM? -> ->Now, what are the chances that FreeBSD Inc could purchase the services ->of Mindcraft to test a properly tuned FreeBSD box vs this NT box? -> ->Stephen. -> -> ->To Unsubscribe: send mail to majordomo@FreeBSD.org ->with "unsubscribe freebsd-hackers" in the body of the message -> -- ____________________________________________________________________ Emmanuel DELOGET [pixel] pixel@{dotcom.fr,epita.fr} ---- DotCom SA -------------------------------------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199904151029.KAA14859>