On Mon, 9 Jun 2003, Stroller wrote:
> On 8/6/03 3:14 pm, "Hemmann, Volker Armin"
> <[EMAIL PROTECTED]> wrote:
>
> > On Sunday 08 June 2003 05:38, Stroller wrote:
> >>
> >> /dev/hda:
> >> Timing buffer-cache reads: 128 MB in 6.86 seconds = 18.66 MB/sec
> >> Timing buffered disk reads: 64 MB in 6.36 seconds = 10.07 MB/sec
> >>...
> >...
> > /dev/hde:
> > Timing buffer-cache reads: 128 MB in 0.42 seconds =304.76 MB/sec
> > Timing buffered disk reads: 64 MB in 5.31 seconds = 12.05 MB/sec
> >...
> > hdg&hde are connected to a ht366 based pci-ide-controller.
> >
> >...
> > hde is an ibm IBM-DTTA-351010 (4 years old?)
> >
> >
> > I doubt that my disks would be any faster with dangerous -X fiddling, but as
> > you can see: your numbers are not so special low... I have a harddisk in a
> > drawer (pio4) that delivers happily 2mb/sec....
>
> Ah..! Ok, I see, now, Volker.
>
> The "Timing buffered disk reads" figure on your IBM drive is about the same
> as mine (within 20% or so).
>
> NOTE TO OTHER POSTERS: your drive isn't slow if it's 10 times as fast as
> mine. I also have a drive 10 times as fast as this one, that one's not the
> point...
>
> But your "Timing buffer-cache reads" figure is fifteen! times as fast as
> mine..! What does this indicate, please..?
>
> Thanks for all your time & help,
>
> Stroller.
hdparm results vary wildly on different drive/controller/ram combinations.
this is a known issue and i don't know if the hdparm people are ever going
to do anything about it. i don't even know why they include the timing
feature.
it's results are totally wacky. ignore the numbers it gives you and do
yourself a favor by just execing an "emerge bonnie".
-j
--
--------------------------------------------------------
Rev. Jeffrey Paul -datavibe- [EMAIL PROTECTED]
aim:x736e65616b pgp:0x15FA257E phone:8777483467
70E0 B896 D5F3 8BF4 4BEE 2CCF EF2F BA28 15FA 257E
--------------------------------------------------------
--
[EMAIL PROTECTED] mailing list
