On Fri, 2004-09-24 at 14:32, James Cuff wrote: > On Fri, 2004-09-24 at 14:21, Dan Bolser wrote: > > On Fri, 24 Sep 2004, elijah wright wrote: > > > Seeker 1...Seeker 3...Seeker 2...start 'em...done...done...done... > > -------Sequential Output-------- ---Sequential Input-- > > --Random-- > > -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- > > --Seeks--- > > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > > 100 4240 25.9 5437 2.1 4780 2.4 19595 100.1 566942 99.7 2023.0 10.1 > > > > With default file size (104857600). The CPU is 'amazingly high' during > > reads. > > I think your file size is to small - 100MB will fit in the UBC. Once > written according to these figures you are seeing ca 600MB/s reads for > getchar ops. The block ops must not be htting the buffer cache > correctly so go back down to 2MB/s. This seems pretty poor even for Doh! That should have been 20MB/s for getchar and 566MB/s for blocks. Sorry - I said I had too much coffee :-) Anyway the take home message is still the same - make the file size at least 2x the amount of memory in your server. Cheers, j.