Am Sa, den 01.05.2004 schrieb Guolin Cheng um 01:31:
> Then that means for normal ia-32 machines, the PCI&IDE Bus speed is
> mostly 33MHz/s, and so the sustainable IDE disk access speed is always
> no more than about 33MB/s? I tried several types of Linux boxes, the
> long-time disk access is about no more than 35MB/s.
Which current IDE hard disk do you expect to have more then 35MB/sec to
50MB/sec sustainable speed? Do not mix maximum possible bus speed with
what you can expect from a drive in real. An remind that manufacturers
overstate with specs given for their products.
> And there are no other means to accelerate disk access speed? Manuauly
> setting IDE disk's access mode (UDMA2/3/4/5/6) with hdparm has no big
> difference, as far as from my own experience.
Normally that really makes no real difference. Only for cases where the
disc can deliver data from their cache.
> Any other ideas, besides buy ia-64 hosts? Thanks.
Why should IA64 change anything? The disk itself is the limiting factor.
> Guolin Cheng
Alexander
--
Alexander Dalloz | Enger, Germany | GPG key 1024D/ED695653 1999-07-13
Fedora GNU/Linux Core 1 (Yarrow) on Athlon CPU kernel 2.4.22-1.2188.nptl
Sirendipity 01:40:16 up 4 days, 29 users, load average: 0.50, 0.28,
[ ÎÎÏÎÎ Ï'ÎÏÏÎÎ - gnothi seauton ]
my life is a planetarium - and you are the stars
Attachment:
signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil
