I have 2 IBM 7200 RPM 80 Gig. DeskStars. Judging by the stats posted on here, my drive performs well. I used the command: hdparm -t /dev/hda.

/dev/hda:
Timing buffered disk reads: 64 MB in 1.41 seconds = 45.39 MB/sec

/dev/hdb:
Timing buffered disk reads: 64 MB in 1.41 seconds = 45.39 MB/sec

I'm using a 5400rpm ATA 66 20.5 G Western Digital and I come up with this:

64 MB in 2.88 seconds = 22.23 MB/sec.

I use a Maxtor Diamond Plus 9, 8Mb.

In my /etc/init.d/rcS (Debian), I have the following line at the top:
hdparm -d1 -X70 -c1 /dev/hda

I ran hdparm test with all my usual programs running (opera, galeon, gcmd, some terminals etc...)

# hdparm -tT /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.60 seconds =213.33 MB/sec
Timing buffered disk reads: 64 MB in 1.35 seconds = 47.41 MB/sec

im using a
western digital ata-66 :
Timing buffered disk reads: 64 MB in 2.30 seconds = 27.83 MB/sec
and western digital ata-66:
Timing buffered disk reads: 64 MB in 2.38 seconds = 26.89 MB/sec

OK, I've read this whole bleedin' thread. I've been wondering about my drive performance even in the new machine I just built.

I'm currently running MDK 9.0 on:
ABIT NF7-2 Ver 1.2 (nForce2)
XP 2200+ Tbred (no OC)
512Mb CAS2 3200 DDR (one stick)
40 GB Maxtor (don't remember the model)
23 GB Maxtor (ditto)
GeForce FX 5200 (twinveiw)
2- 17" CRT

I know both drives are at least UDMA66 capable and I believe the 40 GB is a UDMA 100.

Now, with all that said, and all the reading I've done here today. Do I understand correctly that when I upgrade to MDK 9.1 this week, I won't need to use hdparm because Mandrake will have already optimized it for me?

I'll also be installing MDK 9.1 onto my old machine with a new 80GB WD with 8Mb cache as my home file server.

Has anyone any experience with 9.1 and HD optimization?

Thanks!

Bruce

this post says they have mandrake 9.1

http://www.linuxquestions.org/questi...118#post279118

Did you have to do anything to optimize your hard drives, or did MDK 9.1 just handle it "all by its little self"?

I just got a 600% increase in throughput (3.1 Mb/s to ~24Mb/s) by using hdparm on MDK 9.0.

BTW: Thanks David. I do like the Linux community!

Bruce

System Info:
AMD Thunderbird 1.4 GHz
MSI K7T Pro 266
Seagate 7200RPM 100GB

Results:
root@slackbox:/home/mothugz# hdparm -Tt /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.59 seconds =216.62 MB/sec
Timing buffered disk reads: 64 MB in 1.23 seconds = 52.17 MB/sec

My first hard drive have a *very* poor performance, it's a Maxtor 6Y120P0 120Gb hard drive, making 4.5 Mb/sec while the second one (essentially the same model with less cache, and size) Maxtor 6Y080L0 makes around 50Mb/sec!

I've enabled dma/udma6 (both drives they supports it) and have the same settings for both drives but still, the performace of the 120Gb one is soooo poor!

Here's what I have:

devel:/home/sandor # hdparm -acdgkmnrui /dev/hda

/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 14946/255/63, sectors = 240121728, start = 0

Model=Maxtor 6Y120P0, FwRev=YAR41BW0, SerialNo=Y4180N8E
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=DualPortCache, BuffSize=7936kB, MaxMultSect=16, MultSect=16
CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=240121728
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: (null): 1 2 3 4 5 6 7

devel:/home/sandor # hdparm -t /dev/hda

/dev/hda:
Timing buffered disk reads: 64 MB in 11.76 seconds = 5.44 MB/sec

/dev/hdb:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 9964/255/63, sectors = 160086528, start = 0

Model=Maxtor 6Y080L0, FwRev=YAR41BW0, SerialNo=Y3HLDZSE
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=160086528
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: (null): 1 2 3 4 5 6 7

devel:/home/sandor # hdparm -t /dev/hdb

/dev/hdb:
Timing buffered disk reads: 64 MB in 1.30 seconds = 49.23 MB/sec

Any ideas what shall I do now?

Thanks: Sakov

IBM Travelstar 4200 RPM laptop drive, ATA 100:

Timing buffered disk reads: 64 MB in 3.24 seconds = 19.75 MB/sec

Maxtor Diamonmax 9 Plus 120GB 2MB cache, ATA 33 (old mainboard):

Timing buffered disk reads: 64 MB in 2.52 seconds = 25.40 MB/sec

22.7 MB/sec on an IDE drive.

Not bad, methinks.

Sakov-- no, I do not, as you are encountering in part that the bigger capacity drive needs to move head mosr steps to get to start point, and thus the benchmark is showing the increased latency. Try feeding both drives, one at a time, 2-3 ISOs of data. The first might take longer to start, but for actual writing should be between same and 1.5 times the time of better drive. The benchmark feeds small blocks to various places on hd-- the big ones are bettre for large chunks of data.

Technically, would talk to Maxtor also and see if they can trade you for a pair of smaller HDs unless you want to use the big one for big ISO files and things like that. If, over 1-2 GIG, it actually tiems to take 8-9 times longer, it is poorly designed or not the right drive for your computer or simply defective.

jdii1215, thanks for the iso-copying tip, I've tried to copy a 7G file from one hd to the other; than repeated the test copy from the 2nd to the 1rst hd, In both cases I got an avarage of 26-28Mb/sec copy speed (so taking account the linux caching it is still over 20Mb/sec). So the real throughput is ok I would say.

The wierd thing is this:

Turning on the machine, hdparm -t gives an 28Mb/sec avg. read for the slower 120G drive.
After around 10 mins, it degrades to 20Mb/sec.
After about an hour, it goes down to 10Mb/sec!!!
(I run the test several times to get an avarage)
Needless to say, nothing is running in the background so no process is reading/writing from the drive.

Did anybody experience the same wierd behaviour? The drive is a Maxtor DiamondMax Plus, ATA 133.

Sakov

Western Digital 80 Gb 7200rpm
Timing buffered disk reads: 64 MB in 1.40 seconds = 45.71 MB/sec

Western Digital IDE 13GB 5400 RPM I think ATA 33
Timing buffered disk reads: 64 MB in 2.89 seconds = 22.15 MB/sec

Maxtor IDE 80GB 7200 RPM ATA 133
Timing buffered disk reads: 64 MB in 1.30 seconds = 49.23 MB/sec