On Friday 24 October 2003 16:02, Jesse Keating wrote: > RHL9 Kernel: > > [root kickstation-131 root]# hdparm -Tt /dev/sda > > > /dev/sda: > Timing buffer-cache reads: 2248 MB in 2.00 seconds = 1124.00 > MB/sec Timing buffered disk reads: 196 MB in 3.00 seconds = 65.33 > MB/sec > > [root kickstation-131 data1]# ls -alh bigfile > -rw-r--r-- 1 root root 1000M Oct 23 18:23 bigfile > > [root kickstation-131 data1]# time dd if=bigfile of=secondfile > 2048000+0 records in > 2048000+0 records out > > real 1m2.472s > user 0m1.050s > sys 0m27.900s > > > Interestingly enough, it seems that dd tests are a bit slower in > RHL9, while hdparm is much faster. I'll run tiobench to compare. > I'll also run the tests in a pure RHL9 system. er, scratch these results. One of the workerbees had started CTCS when I started my RHL9 kernel testing, and the results are skewed. Here are the actual RHL9 kernel results: [root kickstation-131 data1]# time dd if=bigfile of=secondfile 2048000+0 records in 2048000+0 records out real 0m37.375s user 0m1.340s sys 0m24.990s Thats the right numbers. So it would definitely seem that the kernel is the culprit in this scenario. RHL9 kernel is using the "megaraid" module by the way, not the megaraid2. -- Jesse Keating RHCE MCSE (geek.j2solutions.net) Fedora Legacy Team (www.fedora.us/wiki/FedoraLegacy) Mondo DevTeam (www.mondorescue.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating
Attachment:
pgp00016.pgp
Description: signature