Date: Mon, 18 Dec 1995 13:12:54 +0100 From: Wolfram Schneider <wosch@cs.tu-berlin.de> To: David Hovemeyer <daveho@infocom.com> Cc: hackers@FreeBSD.ORG Subject: mmap and memory utilization Message-ID: <199512181212.NAA23098@caramba.cs.tu-berlin.de> In-Reply-To: <Pine.BSF.3.91.951216123927.19249G-100000@infocom.com> References: <Pine.BSF.3.91.951216123927.19249G-100000@infocom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
David Hovemeyer writes: >What I am wondering is > > What is the impact of mmap'ing a large file and then > scanning linearly through it? mmap is your friend if the file size is less than 1/2 of memory and if you read the file more than once. Example: locate(1) with mmap and without mmap (fopen, getc) $ ls -l find.codes -rw-rw-r-- 1 wosch 14549501 Dec 18 06:45 find.codes $ uname -s -r SunOS 5.4 (FreeBSD mmap seems a little bit slower than solaris mmap or FreeBSD getc is faster than solaris getc :-) o with mmap $ time ../../bin/sol2/blocate -m -d find.codes woscH real 5.5 user 3.1 sys 0.3 bash$ time ../../bin/sol2/blocate -m -d find.codes woscH real 3.3 user 3.0 sys 0.1 bash$ time ../../bin/sol2/blocate -m -d find.codes woscH real 3.4 user 3.0 sys 0.1 o without mmap $ time ../../bin/sol2/blocate -d find.codes woscH real 6.7 user 4.9 sys 1.1 bash$ time ../../bin/sol2/blocate -d find.codes woscH real 7.0 user 4.9 sys 0.9 bash$ time ../../bin/sol2/blocate -d find.codes woscH real 6.6 user 5.0 sys 0.9
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199512181212.NAA23098>