Date: Tue, 14 Apr 1998 08:48:28 -0700 From: "David E. Tweten" <tweten@frihet.com> To: chad@dcfinc.com Cc: pst@juniper.net, stable@FreeBSD.ORG Subject: Re: MFS in -stable? Message-ID: <199804141548.IAA18006@ns.frihet.com>
next in thread | raw e-mail | index | archive | help
Referring to MFS, chad@freebie.dcfinc.com said: >But if you left the memory available for Kernel buffers, wouldn't you >get the performance without the management headaches? You would get some of the benefit of MFS, if your temporary files were short-lived compared to the sync interval, if they didn't use up too many buffers, and if there weren't too much demand for real memory for processes. You'd still have to endure at least four synchronous disk writes per file for meta data, unless FreeBSD has implemented the recommendations of Ganger and Patt [USENIX SOSDI, 1994]. Their ideas on how to avoid synchronous meta data I/O without endangering file system integrity were definitely not part of 4.4BSD Lite. Synchronous writes to an MFS are _much_ faster than they are to a real disk. -- David E. Tweten | 2047-bit PGP fingerprint: | tweten@frihet.com 12141 Atrium Drive | E9 59 E7 5C 6B 88 B8 90 | tweten@and.com Saratoga, CA 95070-3162 | 65 30 2A A4 A0 BC 49 AE | (408) 446-4131 Those who make good products sell products; those who don't, sell solutions. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199804141548.IAA18006>