Date: Mon, 27 Jan 1997 21:08:56 +1100 From: Bruce Evans <bde@zeta.org.au> To: dicen@hooked.net, freebsd-hackers@FreeBSD.ORG Subject: Re: Performance of ufs vs. ext2. Message-ID: <199701271008.VAA00439@godzilla.zeta.org.au>
next in thread | raw e-mail | index | archive | help
>Okay cool some real numbers. When you speak of "rewrite" are you talking >about the creation and deletion of files (Metadata)? There seams to be a It's whatever the bonnie benchmark does. Something like rewriting the data in an existing file using a too-small blocksize. The small blocksize forces blocks to be written to be pre-read unless the OS is very smart. It is normally at least twice as slow as an append with a too-small blocksize because it involves twice as much i/o. Under FreeBSD it is about 3 times as slow. Under Linux it is only twice as slow. >significant speed difference between the creation and deletion of files >on linux ext2 vs. Freebsd ufs. Linux ext2 is way faster. I suppose I This is caused mainly by a different default for synchronization of writes: FreeBSD: mount -o -noasync: usually safer but always slower. Linux: mount -o -async: usually less safe but always faster. >could just run ext2 under FreeBSD right? It sure would make a "make >world" faster. You know if someone were to setup a news server it would >seam to make more sence to use ext2. No, the file system has very little to do with the speed of create/delete. FreeBSD uses the same mount default for ext2fs and ufs. `mount -o async' speeds them up almost equally so that they are both slightly slower for create/delete than Linux. They are slower because Linux has better support for its default setting. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701271008.VAA00439>