On Thu, 6 Nov 2003 23:58:59 -0500 Kurt Bechstein <[EMAIL PROTECTED]> wrote:
> Does anyone else notice a siginificant difference in cpu usage between > 2.4 and 2.6 kernels when it comes to disk I/O? It seems that 2.6 > really seems to peg the CPU compared to 2.4. I'm using 2.6test9 by the way. > Just curious if this is how things are supposed to be or not. Disk > performance is really good according to hdparm: > /dev/hdb: > Timing buffer-cache reads: 1088 MB in 2.00 seconds = 544.00 MB/sec > Timing buffered disk reads: 134 MB in 3.02 seconds = 44.37 MB/sec > > > Anyone have any thoughts? Thanks. I've never looked on 2.4, but yes cpu utilization bumps to about 76% (Athlon-XP 1800+ 512M) during 'hdparm -tT /dev/hd?' On my Maxtor 30G drive (not a super performer) I get Timing buffer-cache reads: 648 MB in 2.00 seconds = 323.89 MB/sec Timing buffered disk reads: 122 MB in 3.04 seconds = 40.08 MB/sec Kernel 2.6.0_test8 -- Collins Richey - Denver Area if you fill your heart with regrets of yesterday and the worries of tomorrow, you have no today to be thankful for. -- [EMAIL PROTECTED] mailing list
