> which max. sustained read performace can be expected with UDMA33 ?
~33MB/s from buffer to host (- a bit for overheads etc)
> I have a 40GB 7200rpm Maxtor disk which should give min/max 16.7/29.0
MB/sec
> accoring to a benchmark test in german c't magazine.
> but with DFI K6BV3+ main board (MVP3 chipset, only supports UDMA33 not
UDMA66),
> AMD K6/2-400, 128MB PC100 ram, 2MB L2 cache only get ~11-14 MB/sec read
rate.
> I noticed the "maxDMA=0(slow)" in "hdparm -i" output but can't find any
> hint how to change this.
>
> is there any chance to improve/tune performace, or is UDMA33 the real
limit ?
> do I need UDMA66 for 20-30 MB/sec ? why?
Hi Harold,
It looks like you may need a newer version of hdparm to identify the DMA
mode as UDMA2.
I've found the IDE subsystem under linux doesn't perform upto platter speeds
on high-end IDE disks. I have a Seagate Barracuda 20GB, which I get 28.2MB/s
(outer recording zone) under MS Windows 2000, yet 15.3MB/s under linux.
That's almost half!!
I was getting 20.2MB/s with my older Maxtor 5120+ 10GB drive, but I can't
understand why performance is less - although it was a older kernel(!). I
get a nice 128MB/s buffer cache read speed though.
Indeed, I posted a question on these lines, but go no replies. It's a pity,
since that extra speed would be 'free', and satisfy lots of linux (IDE)
speed-merchants.
Dan
__________________________
Daniel J Blueman
Undergraduate - BSc Computing Science
UMIST university - Manchester
Direct line: 0161 933 3569
Mobile: 07775 583766
Email: daniel.j.blueman@stud.umist.ac.uk
SMS: daniel.j.blueman@sms.genie.co.uk
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:19 EST