Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.mel.connect.com.au!news.mel.aone.net.au!grumpy.fl.net.au!news.webspan.net!news.tacom.army.mil!news2.acs.oakland.edu!newsxfer.itd.umich.edu!newsxfer3.itd.umich.edu!howland.erols.net!feed1.news.erols.com!arclight.uoregon.edu!newsfeed.direct.ca!nntp.portal.ca!cynic.portal.ca!not-for-mail From: cjs@cynic.portal.ca (Curt Sampson) Newsgroups: comp.os.linux.misc,comp.os.linux.networking,comp.os.linux.setup,comp.unix.bsd.misc,comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy Subject: Re: Linux vs whatever Date: 29 Jan 1997 23:06:59 -0800 Organization: Internet Portal Services, Inc. Lines: 66 Message-ID: <5cphaj$qvg@cynic.portal.ca> References: <32DFFEAB.7704@usa.net> <5co7vd$lvt@lynx.dac.neu.edu> <5coib1$jvv@cynic.portal.ca> <5cov5p$7hu@lynx.dac.neu.edu> NNTP-Posting-Host: cynic.portal.ca Xref: euryale.cc.adfa.oz.au comp.os.linux.misc:153490 comp.os.linux.networking:65674 comp.os.linux.setup:93243 comp.unix.bsd.misc:1959 comp.os.ms-windows.nt.advocacy:50614 comp.os.os2.advocacy:262825 In article <5cov5p$7hu@lynx.dac.neu.edu>, Michael Kagalenko <mkagalen@lynx.dac.neu.edu> wrote: > As I mentioned, that is legitimate subject to discuss, but this > thread started by allegiation that Linux networking isn't as good as BSDs. No, this thread wasn't started with that. My statement that the quality of BSD ccode is in general considerably better than the quality of NetBSD code was fourteen articles ago. I can see nothing in this thread regarding performance except the position that Linux can excuse worse code with better performance, which is clearly not the case, especially in the example cited, where the performance is demonstrably much worse. > I requested some evidence of it, none of were forthcoming so far. My NFS evidence is available at <http://www2.portal.ca/~cjs/computer/benchmark>. So far all of the stats I've seen from any Linux user have been far below these read benchmarks on any class of machines, and below the write benchmarks on machines with similar CPU power. If you have any evidence to the contrary please present it. Saying `well, there might be evidence to the contrary' without presenting it is like saying `I might toss a ball into the air and it might fall upwards to the sky instead of downwards to the ground.' > In other words you have no evidence, but repeat what everyone else is saying. I do have evidence. I just don't have what I consider to be conclusive evidence. But certainly if Linux's NFS performance is so wonderful, it shouldn't be hard to present some reasonable figures that match my published BSD figures. Let's face it; the BSD evidence is here; the very limited Linux evidence shows that its networking is not as good as BSD's. It's stretches credulity to believe it's some odd coincidence that a) the limited evidence we have shows Linux to be a poor performer, and b) there are all sorts of people out there with better evidence that are reading this discussion and saying `No, I won't post about it.' > This seems to be the most common answer when you press for evidence on > the claims re Linux vs BSD networking. You don't have to press for evidence, you have to produce it. You have a Linux system or two, I presume. Run the tests, rather than complaining that I haven't gone out and installed Linux on my Alpha (presumably after hand-coding a binary patch to the kernel to get around the fact that Linux is too stupid to figure out that it's headless) and done the tests for you. > I am trying > to find some evidence, or references to such evidence, to support either position. No, if you were sincerely trying to find the evidence, you would run some tests and publish the figures. You are trying to weasel your way out of actually admitting that, to the best evidence available at this time, Linux network performance is slower than that of the BSD systems on equivalant hardware. cjs -- Curt Sampson cjs@portal.ca Info at http://www.portal.ca/ Internet Portal Services, Inc. Vancouver, BC (604) 257-9400 De gustibus, aut bene aut nihil.