Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mira.net.au!news.mel.connect.com.au!munnari.OZ.AU!news.ecn.uoknor.edu!news.cis.okstate.edu!newsfeed.ksu.ksu.edu!news.physics.uiowa.edu!math.ohio-state.edu!howland.reston.ans.net!swrinde!newsfeed.internetmci.com!newsxfer2.itd.umich.edu!uunet!in1.uu.net!news.artisoft.com!usenet
From: Terry Lambert <terry@lambert.org>
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: Linux vs. FreeBSD ...
Date: Sat, 11 May 1996 15:26:08 -0700
Organization: Me
Lines: 114
Message-ID: <31951400.600F0009@lambert.org>
References: <3188C1E2.45AE@onramp.net> <4mr1pk$cdi@dyson.iquest.net> <4n0dhd$cff@agate.berkeley.edu> <31940EC3.61419F9C@lambert.org> <4n1hs2$shg@agate.berkeley.edu>
NNTP-Posting-Host: hecate.artisoft.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 2.01 (X11; I; Linux 1.1.76 i486)
Nick Kralevich wrote:
] >NetScape Commerce Server. 8-).
]
] Ya, I had thought about that, but I dismissed it because of the number
] of comparible freely available products that do the same thing.
And because it was a point of the curve you were trying to
prove. Remember that if you want a straight-line fit, you
should only take two data points. 8-)
] And the Netscape Commerce server isn't officially supported by
] Netscape under FreeBSD. If your going to be paying $800+ for the
] commerce server, and not get any support, then that's your choice.
None of the IBCS2 code is "officially supported" on Linux
either. What's your point... that it will run but it's a
bad moral decision to run it?
] (oh no, not the sync vs async metadata argument again. Ahhhhh!!!)
]
] Perhaps that's the difference between scientists and engineers. This
] is just another example of the risk/rewards payoff. For volumes
] mounted async, there is approximately 10X increase in speed vs
] async for operations which do a lot of metadata update (such
] as untaring a file), with a minimal (some would say non-existant)
] increase in risk.
1) It's "doing a lot of directory manipulations" that speeds
up, not "untarring a file".
2) It's "some idiots would claim non-existant", not "some
would say non-existant". It's on the order of salemen
making technical claims on sound cards.
] As you've pointed out, FreeBSD supports async writes. As you didn't
] point out, Linux supports sync writes. Both options are available
] on both systems.
This point has been (over) made before. It's the defaults that
exist without user intervention that are an important indicator
of philosophy.
] FreeBSD comes with sync writes on by default, where Linux has async
] writes by default. I would argue that the Linux default is more
] flexable than the FreeBSD default. Under the default Linux setup,
] applications requiring sync access to the disk can always call
] "fsync()", "fdatasync()", or "msync()", or open the file with the
] "O_SYNC" option.
]
] However, with FreeBSD defaulting to sync updates, it is impossible
] to get async write behavior out of the filesystem. The Linux
] default is more flexable and offers higher performance at a minor
] (insignificant?) increase in risk.
Clearly you do not understand the issue. Rather than correct you
at length, I will simply note that "fsync()", "fdatasync()",
"msync()", and the "O_SYNC" option do not affect handling of
metadata, and so are irrelevant to any discussion of the subject.
Your statement about FreeBSD, that "it is impossible to get async
write behavior out of the filesystem", is simply false.
] It's up to the individual user to make the choice to take or reject
] the risk vs reward of async updates. Lots of users take risks that
] are much worse than this, such as running without a UPS or not making
] backups.
The risks should not be accepted on their behalf by default.
] >] Oh, one more quick question for FreeBSDers. Does FreeBSD support
] >] POSIX 1b realtime extentions yet? Is anyone working on it? This
] >] is a serious question.
] >
] >Yes. Patches available on request to the author on the
] >FreeBSD RealTime list.
]
] Boy, you folks don't go out of your way to advertise this list, do
] you? The list is not searchable in the FreeBSD search page
] (http://www.freebsd.org/search.html),
It's not archived. Only archived lists are searchable.
] is not described in the FreeBSD handbook under "mailing lists"
] (http://www.freebsd.org/handbook/handbook251.html#459), nor is
] it described anywhere on the FreeBSD WWW site.
This is incorrect. I leave it to you to find where, however.
Note that it is one of the lists returned as available by
majordomo; read the FAQ.
] But thank you for the information. I have an application I'm
] developing (a parallel MPEG encoder), that would benefit from the
] "aio_read()" and "aio_write()" syscalls, and I would like to see
] how (or if) FreeBSD impliments them. I've sent e-mail to the
] realtime list asking them to send me the location of the patches.
aio_read/aio_write are not RealTime extensions.
] >Does Linux support v10 of the POSIX threading, or only v4 of
] >1001.3b? This is a serious question.
]
] I'll do some research and get back to you on this.
I'll save you the trouble; it's v4.
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.