ECS K7S5A: IDE performance

ECS K7S5A: IDE performance

[Frédéric L. W. Meunier]

I have an ECS K7S5A 3.1A. It works fine with 2.4.19. No
corruption. Now I tested it with hdparm and:

hdparm -tT /dev/hda

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.79 seconds =162.03 MB/sec
 Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec

Only 38.10 ?

hdparm -i /dev/hda

/dev/hda:

 Model=MAXTOR 6L060J3, FwRev=A93.0500, SerialNo=663200252994
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
 BuffType=DualPortCache, BuffSize=1819kB, MaxMultSect=16, MultSect=off
 CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=117266688
 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=no WriteCache=enabled
 Drive conforms to: ATA/ATAPI-5 T13 1321D revision 1:  1 2 3 4 5

SIS5513: IDE controller on PCI bus 00 dev 15
SIS5513: chipset revision 208
SIS5513: not 100% native mode: will probe irqs later
SiS735    ATA 100 controller
    ide0: BM-DMA at 0xff00-0xff07, BIOS settings: hda:DMA, hdb:DMA
    ide1: BM-DMA at 0xff08-0xff0f, BIOS settings: hdc:DMA, hdd:DMA
hda: MAXTOR 6L060J3, ATA DISK drive

hda: 117266688 sectors (60041 MB) w/1819KiB Cache, CHS=7299/255/63, UDMA(100)

-- 
0@pervalidus.{net, {dyndns.}org} ICQ: 156552241
Tel: +55 (21) 2717-2399 (with voice mailbox)

Re: ECS K7S5A: IDE performance

[Tomas Szepe]

> I have an ECS K7S5A 3.1A. It works fine with 2.4.19. No
> corruption. Now I tested it with hdparm and:
> 
> hdparm -tT /dev/hda
> 
> /dev/hda:
>  Timing buffer-cache reads:   128 MB in  0.79 seconds =162.03 MB/sec
>  Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec
> 
> Only 38.10 ?

How do you mean, only 38.10?

> hdparm -i /dev/hda
> 
> /dev/hda:
> 
>  Model=MAXTOR 6L060J3, FwRev=A93.0500, SerialNo=663200252994
>  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
>  RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
>  BuffType=DualPortCache, BuffSize=1819kB, MaxMultSect=16, MultSect=off
>  CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=117266688
>  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=no WriteCache=enabled
>  Drive conforms to: ATA/ATAPI-5 T13 1321D revision 1:  1 2 3 4 5

Looks ok to me.

T.

Re: ECS K7S5A: IDE performance

[Vojtech Pavlik]

On Wed, Aug 28, 2002 at 04:00:26PM -0300, Fr?d?ric L. W. Meunier wrote:

> I have an ECS K7S5A 3.1A. It works fine with 2.4.19. No
> corruption. Now I tested it with hdparm and:
> 
> hdparm -tT /dev/hda
> 
> /dev/hda:
>  Timing buffer-cache reads:   128 MB in  0.79 seconds =162.03 MB/sec
>  Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec
> 
> Only 38.10 ?

That's quite a LOT for an IDE drive.

> hdparm -i /dev/hda
> 
> /dev/hda:
> 
>  Model=MAXTOR 6L060J3, FwRev=A93.0500, SerialNo=663200252994
>  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
>  RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
>  BuffType=DualPortCache, BuffSize=1819kB, MaxMultSect=16, MultSect=off
>  CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=117266688
>  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=no WriteCache=enabled
>  Drive conforms to: ATA/ATAPI-5 T13 1321D revision 1:  1 2 3 4 5
> 
> SIS5513: IDE controller on PCI bus 00 dev 15
> SIS5513: chipset revision 208
> SIS5513: not 100% native mode: will probe irqs later
> SiS735    ATA 100 controller
>     ide0: BM-DMA at 0xff00-0xff07, BIOS settings: hda:DMA, hdb:DMA
>     ide1: BM-DMA at 0xff08-0xff0f, BIOS settings: hdc:DMA, hdd:DMA
> hda: MAXTOR 6L060J3, ATA DISK drive
> 
> hda: 117266688 sectors (60041 MB) w/1819KiB Cache, CHS=7299/255/63, UDMA(100)
> 
> -- 
> 0@pervalidus.{net, {dyndns.}org} ICQ: 156552241
> Tel: +55 (21) 2717-2399 (with voice mailbox)
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
Vojtech Pavlik
SuSE Labs

Re: ECS K7S5A: IDE performance

[Frédéric L. W. Meunier]

On Wed, 28 Aug 2002, Tomas Szepe wrote:

> > I have an ECS K7S5A 3.1A. It works fine with 2.4.19. No
> > corruption. Now I tested it with hdparm and:
> >
> > hdparm -tT /dev/hda
> >
> > /dev/hda:
> >  Timing buffer-cache reads:   128 MB in  0.79 seconds =162.03 MB/sec
> >  Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec
> >
> > Only 38.10 ?
>
> How do you mean, only 38.10?

I just thought it'd be much more with an ATA100. I got more or
less the same with my earlier motherboard, an ASUS A7APro, and
without ATA66 - which would print a lot of CRC errors at boot
time if enabled in the BIOS. The K7S5A doesn't print any and is
rock solid.

Maybe running it at 100/133 (and not 100/100) decreases
performance ? I read is somewhere. I have an Athlon 1000 (200)
with 2x256Mb DDR PC2100.

-- 
0@pervalidus.{net, {dyndns.}org} ICQ: 156552241
Tel: +55 (21) 2717-2399 (with voice mailbox)

Re: ECS K7S5A: IDE performance

[Vojtech Pavlik]

On Wed, Aug 28, 2002 at 04:19:13PM -0300, Fr?d?ric L. W. Meunier wrote:
> On Wed, 28 Aug 2002, Tomas Szepe wrote:
> 
> > > I have an ECS K7S5A 3.1A. It works fine with 2.4.19. No
> > > corruption. Now I tested it with hdparm and:
> > >
> > > hdparm -tT /dev/hda
> > >
> > > /dev/hda:
> > >  Timing buffer-cache reads:   128 MB in  0.79 seconds =162.03 MB/sec
> > >  Timing buffered disk reads:  64 MB in  1.68 seconds = 38.10 MB/sec
> > >
> > > Only 38.10 ?
> >
> > How do you mean, only 38.10?
> 
> I just thought it'd be much more with an ATA100. I got more or
> less the same with my earlier motherboard, an ASUS A7APro, and
> without ATA66 - which would print a lot of CRC errors at boot
> time if enabled in the BIOS. The K7S5A doesn't print any and is
> rock solid.
> 
> Maybe running it at 100/133 (and not 100/100) decreases
> performance ? I read is somewhere. I have an Athlon 1000 (200)
> with 2x256Mb DDR PC2100.

No. The limit is the drive platter read speed.

-- 
Vojtech Pavlik
SuSE Labs

Re: ECS K7S5A: IDE performance

[Chris Friesen]

Frédéric L. W. Meunier wrote:

> I just thought it'd be much more with an ATA100. I got more or
> less the same with my earlier motherboard, an ASUS A7APro, and
> without ATA66 - which would print a lot of CRC errors at boot
> time if enabled in the BIOS. The K7S5A doesn't print any and is
> rock solid.

If you're reading more than the size of your on-drive buffer then you'll be limited by the speed at
which the information can be grabbed off the drive--in your case, 38.1MB/s, which is quite good.

ATA33/66/100/133 only makes a significant difference in speed when you're reading from the on-drive
cache.

Chris

Re: ECS K7S5A: IDE performance

[Daniel Egger]

[-- Attachment #1: Type: text/plain, Size: 269 bytes --]

Am Mit, 2002-08-28 um 21.19 schrieb Frédéric L. W. Meunier:

> I just thought it'd be much more with an ATA100.

Wellcome to a world where features are sold by marketing.
ATA133, 2.8GHz, 200 Watts minispeakers, you get the idea....

-- 
Servus,
       Daniel

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: ECS K7S5A: IDE performance

[Lionel Bouton]

Chris Friesen wrote:

>Frédéric L. W. Meunier wrote:
>
>>I just thought it'd be much more with an ATA100. I got more or
>>less the same with my earlier motherboard, an ASUS A7APro, and
>>without ATA66 - which would print a lot of CRC errors at boot
>>time if enabled in the BIOS. The K7S5A doesn't print any and is
>>rock solid.
>>
>
>If you're reading more than the size of your on-drive buffer then you'll be limited by the speed at
>which the information can be grabbed off the drive--in your case, 38.1MB/s, which is quite good.
>
>ATA33/66/100/133 only makes a significant difference in speed when you're reading from the on-drive
>cache.
>

In fact I remember having seen something like 10-20% throughput increase 
for the same drive used in UDMA mode 4 (66MB/s) on an HPT366 and UDMA 
mode 5 (100MB/s) on a SiS735 for a drive which maxed out at around 
30/35MB/s.

This could be either caused by IDE controller/PCI interface 
(mis)behaviour, drive firmware tuned for ATA100 instead of ATA66, IDE 
bus overheads. I don't know for sure, but apparently switching from 
ATA66 to ATA100 can bring some juice in some cases where the max platter 
throughput is not even near max advertised IDE bandwidth.

Anyway, to answer to the original post, 38.1 MB/s is quite good.

LB.

Re: ECS K7S5A: IDE performance

[Dieter Nützel]

Chris Friesen wrote:
> Frédéric L. W. Meunier wrote:
>
> > I just thought it'd be much more with an ATA100. I got more or
> > less the same with my earlier motherboard, an ASUS A7APro, and
> > without ATA66 - which would print a lot of CRC errors at boot
> > time if enabled in the BIOS. The K7S5A doesn't print any and is
> > rock solid.
>
> If you're reading more than the size of your on-drive buffer then you'll be
> limited by the speed at which the information can be grabbed off the
> drive--in your case, 38.1MB/s, which is quite good.

IBM (Yes, I konw...;-) IC35L060AVER07 (the whole family) is at 45,6 MB/s.
Maxtor 80 GB nearly the same.

System: AMD 750 (ATA4) and 760 MPX (ATA5).

> ATA33/66/100/133 only makes a significant difference in speed when you're
> reading from the on-drive cache.

Or maybe ATA RAID...;-)

Regards,
	Dieter

-- 
Dieter Nützel
Graduate Student, Computer Science

University of Hamburg
Department of Computer Science
@home: Dieter.Nuetzel at hamburg.de (replace at with @)