Date: Sat, 25 Sep 1999 15:18:10 -0700 (PDT) From: Alfred Perlstein <bright@wintelcom.net> To: john <jbarbee@singular.com> Cc: freebsd-questions@FreeBSD.ORG Subject: Re: performance increase with MSF? Message-ID: <Pine.BSF.4.05.9909251515430.6368-100000@fw.wintelcom.net> In-Reply-To: <4.2.0.58.19990925130141.00d75370@server0.singular.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 25 Sep 1999, john wrote: > server# df -k > Filesystem 1K-blocks Used Avail Capacity Mounted on > .... > procfs 4 4 0 100% /proc > mfs:337 7903 7234 37 99% /usr/local/www/global/tmp > > testfile is a 7 meg file created from a couple of tar files. > the 7234K used are taken up by testfile. there are no other files in the > mfs filesystem. > > server# /usr/bin/time grep blah testfile > ...grep output... > 0.11 real 0.06 user 0.05 sys > server# /usr/bin/time grep blah /usr/local/www/global/tmp/testfile > ...grep output... > 0.11 real 0.06 user 0.05 sys > > Is my method of testing invalid? Is 7 megs too small to make a > difference? Did I miss something when I read the docs? Yes, basically when you create the file it's in the buffercache you aren't doing enough to clear out the cache before you run your tests and both tests are seeing the results of cached file accesses. You need to purge the cache by reading a lot of data from unrelated files. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.05.9909251515430.6368-100000>