Re: How does on get a HD running at a descent speed

Zlatko Calusic (Zlatko.Calusic@CARNet.hr)
25 Nov 1997 08:45:49 +0100


jan gyselinck <JAN.GYSELINCK@student.kuleuven.ac.be> writes:

> Hi,
>
> Maybe this is not something to post here (sounds familiar), if not,
> it's just some distraction.
>
> I got a notebook, a Mitac 5021, with an 516MB IBM harddisk, wich seems
> very slow. When I test it with hdparm, i get something like 1.1 MB/sec,
> which is not much, not much at all. I tried fiddeling with hdparm 3.1,
> with little (0.1MB more or less per second) success. It is also very
> annoying that writing a little bit to disc interrupts PLIP, SLIP and ppa
> transfers (yes, that's normal, but it would be less if this HD would
> work faster)
>
> I have a driver for DOS which puts this drive in PIO mode 3, but this
> doesn't change anything when I run it before loadlin. Under win95, the
> drivers don't work at all, but it would be nice to have some speed under
> linux. I tried that patch for generic DMA support, but that doesn't make
> any difference.
>
> If anyone knows how I could get some descent speed with my HD,
> say it (I read the list, no need to CC to me).
>

My magic recipe is:

hdparm -u1 -m16 -a16 -A1 -c1 /dev/hda

Though your mileage may vary, of course.

Improvements I got:

-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
100 2883 94.1 3868 66.0 1566 78.1 2151 84.3 3288 86.6 52.1 9.0
+hdparm 100 3819 90.3 6100 51.4 1762 39.8 2821 82.7 4860 69.6 49.6 4.4

Notice how %CPU columns have lower values, even with better I/O
throughput. Tested disk is IBM DAQA33240 (3.2GB).

Good luck!

-- 
Posted by Zlatko Calusic           E-mail: <Zlatko.Calusic@CARNet.hr>
---------------------------------------------------------------------
		  So much time, and so little to do.