* Re: disk transfer speed problem
[not found] <Pine.LNX.4.33.0202271701360.13103-100000@coffee.psychology.mcmaster.ca>
@ 2002-02-27 22:52 ` lonely wolf
2002-02-27 23:46 ` Alan Cox
2002-03-01 10:51 ` Marco Colombo
0 siblings, 2 replies; 15+ messages in thread
From: lonely wolf @ 2002-02-27 22:52 UTC (permalink / raw)
To: Mark Hahn, linux-kernel
Mark Hahn wrote:
> > initial data: one brand new intel bonham motherboard (i815) , 900 MHz
> > celeron processor
>
> what's the ram config? that's probably one of the old/slow
> FSB66 celerons, but is the ram fast?
the processor has 100 MB FSB
The memory is PC133, 1 DIMM of 128 MB
> > DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
>
> fine, udma100.
i know
> > The problem:
> > #/sbin/hdparm -tTf /dev/hdd
> >
> > /dev/hdd:
> > Timing buffer-cache reads: 128 MB in 1.17 seconds =109.40 MB/sec
> > Timing buffered disk reads: 64 MB in 3.24 seconds = 19.75 MB/sec
>
> well, 109 MB/s is pretty low for buffer-cache reads; this reflects
> the relative crippled-ness of your cpu/dram/chipset.
well... i would't name a Celeron 900 MHz crippled. PC133 is the best the
board gets... and now the speed is lower then the previous server which was
an Athlon 600 pluggede in an Asus VIA KX133 based mobo.
> doesn't the i815 have integrated video,
yes
> and if so, are you using it,
yes
> and if so, do you have the optional "display cache"?
couldn't find any BIOS setting for that. where should it be ?
> without the DC, video steals quite a lot of dram bandwidth...
the machine is (should be) a NFS server, I couldn't care less about the
video
--
Manuel Wolfshant linux registered user #131416
network administrator NoBug Consulting Romania
Beware the fury of a patient man.
^ permalink raw reply [flat|nested] 15+ messages in thread* Re: disk transfer speed problem
2002-02-27 22:52 ` disk transfer speed problem lonely wolf
@ 2002-02-27 23:46 ` Alan Cox
2002-02-28 9:20 ` Davidovac Zoran
2002-03-01 2:25 ` lonely wolf
2002-03-01 10:51 ` Marco Colombo
1 sibling, 2 replies; 15+ messages in thread
From: Alan Cox @ 2002-02-27 23:46 UTC (permalink / raw)
To: lonely wolf; +Cc: Mark Hahn, linux-kernel
> > > Timing buffered disk reads: 64 MB in 3.24 seconds = 19.75 MB/sec
> >
> > well, 109 MB/s is pretty low for buffer-cache reads; this reflects
> > the relative crippled-ness of your cpu/dram/chipset.
>
> well... i would't name a Celeron 900 MHz crippled. PC133 is the best the
> board gets... and now the speed is lower then the previous server which was
> an Athlon 600 pluggede in an Asus VIA KX133 based mobo.
I get 25MB/sec off my i815 board. It is pretty starved - I seem stuck at
about 25MB/sec total even doing hdparm across both controllers.
Using an external video card might make a small difference
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-27 23:46 ` Alan Cox
@ 2002-02-28 9:20 ` Davidovac Zoran
2002-02-28 14:49 ` lonely wolf
2002-03-01 2:25 ` lonely wolf
1 sibling, 1 reply; 15+ messages in thread
From: Davidovac Zoran @ 2002-02-28 9:20 UTC (permalink / raw)
To: Alan Cox; +Cc: lonely wolf, Mark Hahn, linux-kernel
On Wed, 27 Feb 2002, Alan Cox wrote:
> Date: Wed, 27 Feb 2002 23:46:42 +0000 (GMT)
> From: Alan Cox <alan@lxorguk.ukuu.org.uk>
> To: lonely wolf <wolfy@pcnet.ro>
> Cc: Mark Hahn <hahn@physics.mcmaster.ca>, linux-kernel@vger.kernel.org
>
> > > > Timing buffered disk reads: 64 MB in 3.24 seconds = 19.75 MB/sec
> > >
> > > well, 109 MB/s is pretty low for buffer-cache reads; this reflects
> > > the relative crippled-ness of your cpu/dram/chipset.
> >
> > well... i would't name a Celeron 900 MHz crippled. PC133 is the best the
> > board gets... and now the speed is lower then the previous server which was
> > an Athlon 600 pluggede in an Asus VIA KX133 based mobo.
>
> I get 25MB/sec off my i815 board. It is pretty starved - I seem stuck at
> about 25MB/sec total even doing hdparm across both controllers.
>
> Using an external video card might make a small difference
> -
on kernel 2.2.20 with andre ata100 patches ported from 2.2.19 to 2.2.20
on asus i815 board.
Kernel 2.4.17 results are similar about 35MB/sec
I got this results, machine is under small workload.
root@MEGANET:~# hdparm -tT /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 0.96 seconds =133.33 MB/sec
Timing buffered disk reads: 64 MB in 1.73 seconds = 36.99 MB/sec
root@MEGANET:~# hdparm -i /dev/hda
/dev/hda:
Model=MAXTOR 6L040J2, FwRev=A93.0500, SerialNo=662131534628
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=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=78177792
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 udma0 udma1 udma2 udma3 udma4 *udma5 udma6
AdvancedPM=no
Drive Supports : ATA/ATAPI-5 T13 1321D revision 1 : ATA-1 ATA-2 ATA-3
ATA-4
ATA-5
root@MEGANET:~# hdparm -v /dev/hda
/dev/hda:
multcount = 0 (off)
I/O support = 0 (default 16-bit)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 4866/255/63, sectors = 78177792, start = 0
other details:
Intel machine check reporting enabled on CPU#0.
128K L2 cache (4 way)
CPU: L2 Cache: 128K
CPU: Intel Pentium III (Coppermine) stepping 0a
Checking 386/387 coupling... OK, FPU using exception 16 error reporting.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.35a (19990819) Richard Gooch (rgooch@atnf.csiro.au)
PCI: PCI BIOS revision 2.10 entry at 0xf0e30
PCI: Using configuration type 1
PCI: Probing PCI hardware
Linux agpgart interface v0.99 (c) Jeff Hartmann
agpgart: Maximum main memory to use for agp memory: 203M
agpgart: Detected Intel i815 chipset
agpgart: AGP aperture is 32M @ 0xfc000000
Zoran
^ permalink raw reply [flat|nested] 15+ messages in thread* Re: disk transfer speed problem
2002-02-28 9:20 ` Davidovac Zoran
@ 2002-02-28 14:49 ` lonely wolf
0 siblings, 0 replies; 15+ messages in thread
From: lonely wolf @ 2002-02-28 14:49 UTC (permalink / raw)
To: linux-kernel
Davidovac Zoran wrote:
> > -
> on kernel 2.2.20 with andre ata100 patches ported from 2.2.19 to 2.2.20
> on asus i815 board.
> Kernel 2.4.17 results are similar about 35MB/sec
Any chance to get better results with stock kernel 2.4.17 then with 2.4.9-21 (or
-31) ?
I am not in position to do many tests (or reboots for that matter), the machine is
in production, so I'd like a definite approach, if at all possible
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-27 23:46 ` Alan Cox
2002-02-28 9:20 ` Davidovac Zoran
@ 2002-03-01 2:25 ` lonely wolf
2002-03-01 10:59 ` Andre Hedrick
1 sibling, 1 reply; 15+ messages in thread
From: lonely wolf @ 2002-03-01 2:25 UTC (permalink / raw)
To: linux-kernel
From: Tim Moore
>Please post dmesg, /proc/pci and chipset info from /proc/ide. Also,
>what is result from 'hdparm -tT /dev/hda'?
From: Andre Hedrick
>What is more useful is the cat /proc/ide/ide0/config !!!
[root@bugmaster /root]# cat /proc/ide/ide0/config
pci bus 00 device f9 vid 8086 did 244b channel 0
86 80 4b 24 05 00 80 02 11 80 01 01 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
a1 ff 00 00 00 00 00 00 00 00 00 00 86 80 4d 42
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
07 a3 77 e3 38 00 00 00 0d 00 02 33 00 00 00 00
00 00 00 00 5d 04 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 4d 04 00 00 00 00 00 00
[root@bugmaster /root]# cat /proc/ide/ide1/config
pci bus 00 device f9 vid 8086 did 244b channel 1
86 80 4b 24 05 00 80 02 11 80 01 01 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
a1 ff 00 00 00 00 00 00 00 00 00 00 86 80 4d 42
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
07 a3 77 e3 38 00 00 00 0d 00 02 33 00 00 00 00
00 00 00 00 5d 04 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 4d 04 00 00 00 00 00 00
[root@bugmaster /root]# cat /proc/pci
PCI devices found:
Bus 0, device 0, function 0:
Host bridge: PCI device 8086:1130 (Intel Corporation) (rev 4).
Bus 0, device 2, function 0:
VGA compatible controller: PCI device 8086:1132 (Intel Corporation) (rev 4).
IRQ 11.
Prefetchable 32 bit memory at 0xf8000000 [0xfbffffff].
Non-prefetchable 32 bit memory at 0xffa80000 [0xffafffff].
Bus 0, device 30, function 0:
PCI bridge: Intel Corporation 82801BAM PCI (rev 17).
Master Capable. No bursts. Min Gnt=6.
Bus 0, device 31, function 0:
ISA bridge: Intel Corporation 82801BA ISA Bridge (ICH2) (rev 17).
Bus 0, device 31, function 1:
IDE interface: Intel Corporation 82801BA IDE U100 (rev 17).
I/O at 0xffa0 [0xffaf].
Bus 0, device 31, function 2:
USB Controller: Intel Corporation 82801BA(M) USB (Hub A) (rev 17).
IRQ 10.
I/O at 0xef40 [0xef5f].
Bus 0, device 31, function 3:
SMBus: Intel Corporation 82801BA(M) SMBus (rev 17).
IRQ 9.
I/O at 0xefa0 [0xefaf].
Bus 0, device 31, function 4:
USB Controller: Intel Corporation 82801BA(M) USB (Hub B) (rev 17).
IRQ 6.
I/O at 0xef80 [0xef9f].
Bus 1, device 8, function 0:
Ethernet controller: Intel Corporation 82801BA(M) Ethernet (rev 3).
IRQ 3.
Master Capable. Latency=32. Min Gnt=8.Max Lat=56.
Non-prefetchable 32 bit memory at 0xff8ff000 [0xff8fffff].
I/O at 0xdf00 [0xdf3f].
[root@bugmaster /root]# cat /proc/ide/drivers
ide-floppy version 0.98
ide-disk version 1.11
[root@bugmaster /root]# cat /proc/ide/piix
Intel PIIX4 Ultra 100 Chipset.
--------------- Primary Channel ---------------- Secondary Channel -------------
enabled enabled
--------------- drive0 --------- drive1 -------- drive0 ---------- drive1 ------
DMA enabled: yes no yes yes
UDMA enabled: yes no yes yes
UDMA enabled: 4 X 4 4
UDMA
DMA
PIO
[root@bugmaster /root]# /sbin/hdparm -tT /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 1.10 seconds =116.36 MB/sec
Timing buffered disk reads: 64 MB in 3.73 seconds = 17.16 MB/sec
[root@bugmaster /root]# /sbin/hdparm -tT /dev/hda;/sbin/hdparm -i /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 1.11 seconds =115.32 MB/sec
Timing buffered disk reads: 64 MB in 3.48 seconds = 18.39 MB/sec
/dev/hda:
Model=QUANTUM FIREBALLlct15 07, FwRev=A01.0F00, SerialNo=611025994167
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=15522/15/63, TrkSize=32256, SectSize=21298, ECCbytes=4
BuffType=DualPortCache, BuffSize=418kB, MaxMultSect=16, MultSect=16
CurCHS=15522/15/63, CurSects=-771620641, LBA=yes, LBAsects=14668290
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 udma0 udma1 udma2 udma3 *udma4
^ permalink raw reply [flat|nested] 15+ messages in thread* Re: disk transfer speed problem
2002-03-01 2:25 ` lonely wolf
@ 2002-03-01 10:59 ` Andre Hedrick
0 siblings, 0 replies; 15+ messages in thread
From: Andre Hedrick @ 2002-03-01 10:59 UTC (permalink / raw)
To: lonely wolf; +Cc: linux-kernel
On Fri, 1 Mar 2002, lonely wolf wrote:
> From: Tim Moore
> >Please post dmesg, /proc/pci and chipset info from /proc/ide. Also,
> >what is result from 'hdparm -tT /dev/hda'?
>
> From: Andre Hedrick
> >What is more useful is the cat /proc/ide/ide0/config !!!
>
Well it is clear now, after obtaining the unpublished errata that a minor
bug may be showing it face. It has to do with the programable timing
index and which registers can be set. It also states that multiple bit
sets shall default to the highest supported clock.
So I will need to look at the docs public, unpublished, and call my friend
again to verify.
IIRC, he stated there is no problem but I am always paranoid about timing
issues.
Cheers,
> [root@bugmaster /root]# cat /proc/ide/ide0/config
> pci bus 00 device f9 vid 8086 did 244b channel 0
> 86 80 4b 24 05 00 80 02 11 80 01 01 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> a1 ff 00 00 00 00 00 00 00 00 00 00 86 80 4d 42
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 07 a3 77 e3 38 00 00 00 0d 00 02 33 00 00 00 00
> 00 00 00 00 5d 04 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 4d 04 00 00 00 00 00 00
>
> [root@bugmaster /root]# cat /proc/ide/ide1/config
> pci bus 00 device f9 vid 8086 did 244b channel 1
> 86 80 4b 24 05 00 80 02 11 80 01 01 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> a1 ff 00 00 00 00 00 00 00 00 00 00 86 80 4d 42
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 07 a3 77 e3 38 00 00 00 0d 00 02 33 00 00 00 00
> 00 00 00 00 5d 04 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 00 00 00 00 00 00 00 00 4d 04 00 00 00 00 00 00
>
> [root@bugmaster /root]# cat /proc/pci
> PCI devices found:
> Bus 0, device 0, function 0:
> Host bridge: PCI device 8086:1130 (Intel Corporation) (rev 4).
> Bus 0, device 2, function 0:
> VGA compatible controller: PCI device 8086:1132 (Intel Corporation) (rev 4).
>
> IRQ 11.
> Prefetchable 32 bit memory at 0xf8000000 [0xfbffffff].
> Non-prefetchable 32 bit memory at 0xffa80000 [0xffafffff].
> Bus 0, device 30, function 0:
> PCI bridge: Intel Corporation 82801BAM PCI (rev 17).
> Master Capable. No bursts. Min Gnt=6.
> Bus 0, device 31, function 0:
> ISA bridge: Intel Corporation 82801BA ISA Bridge (ICH2) (rev 17).
> Bus 0, device 31, function 1:
> IDE interface: Intel Corporation 82801BA IDE U100 (rev 17).
> I/O at 0xffa0 [0xffaf].
> Bus 0, device 31, function 2:
> USB Controller: Intel Corporation 82801BA(M) USB (Hub A) (rev 17).
> IRQ 10.
> I/O at 0xef40 [0xef5f].
> Bus 0, device 31, function 3:
> SMBus: Intel Corporation 82801BA(M) SMBus (rev 17).
> IRQ 9.
> I/O at 0xefa0 [0xefaf].
> Bus 0, device 31, function 4:
> USB Controller: Intel Corporation 82801BA(M) USB (Hub B) (rev 17).
> IRQ 6.
> I/O at 0xef80 [0xef9f].
> Bus 1, device 8, function 0:
> Ethernet controller: Intel Corporation 82801BA(M) Ethernet (rev 3).
> IRQ 3.
> Master Capable. Latency=32. Min Gnt=8.Max Lat=56.
> Non-prefetchable 32 bit memory at 0xff8ff000 [0xff8fffff].
> I/O at 0xdf00 [0xdf3f].
>
> [root@bugmaster /root]# cat /proc/ide/drivers
> ide-floppy version 0.98
> ide-disk version 1.11
>
> [root@bugmaster /root]# cat /proc/ide/piix
>
> Intel PIIX4 Ultra 100 Chipset.
> --------------- Primary Channel ---------------- Secondary Channel -------------
> enabled enabled
> --------------- drive0 --------- drive1 -------- drive0 ---------- drive1 ------
> DMA enabled: yes no yes yes
> UDMA enabled: yes no yes yes
> UDMA enabled: 4 X 4 4
> UDMA
> DMA
> PIO
>
> [root@bugmaster /root]# /sbin/hdparm -tT /dev/hda
>
> /dev/hda:
> Timing buffer-cache reads: 128 MB in 1.10 seconds =116.36 MB/sec
> Timing buffered disk reads: 64 MB in 3.73 seconds = 17.16 MB/sec
> [root@bugmaster /root]# /sbin/hdparm -tT /dev/hda;/sbin/hdparm -i /dev/hda
>
> /dev/hda:
> Timing buffer-cache reads: 128 MB in 1.11 seconds =115.32 MB/sec
> Timing buffered disk reads: 64 MB in 3.48 seconds = 18.39 MB/sec
>
> /dev/hda:
>
> Model=QUANTUM FIREBALLlct15 07, FwRev=A01.0F00, SerialNo=611025994167
> Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
> RawCHS=15522/15/63, TrkSize=32256, SectSize=21298, ECCbytes=4
> BuffType=DualPortCache, BuffSize=418kB, MaxMultSect=16, MultSect=16
> CurCHS=15522/15/63, CurSects=-771620641, LBA=yes, LBAsects=14668290
> 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 udma0 udma1 udma2 udma3 *udma4
>
>
>
> -
> 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/
>
Andre Hedrick
Linux Disk Certification Project Linux ATA Development
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-27 22:52 ` disk transfer speed problem lonely wolf
2002-02-27 23:46 ` Alan Cox
@ 2002-03-01 10:51 ` Marco Colombo
1 sibling, 0 replies; 15+ messages in thread
From: Marco Colombo @ 2002-03-01 10:51 UTC (permalink / raw)
To: lonely wolf; +Cc: linux-kernel
On Thu, 28 Feb 2002, lonely wolf wrote:
> Mark Hahn wrote:
[...]
> > well, 109 MB/s is pretty low for buffer-cache reads; this reflects
> > the relative crippled-ness of your cpu/dram/chipset.
>
> well... i would't name a Celeron 900 MHz crippled. PC133 is the best the
> board gets... and now the speed is lower then the previous server which was
> an Athlon 600 pluggede in an Asus VIA KX133 based mobo.
I don't really see why you're expecting a performance breakthru downgrading
from Athlon to Celeron... many all-in-one MB (targeted to Celeron system)
are somewhat crippled and CPU MHz mean close to nothing for I/O performance
(or performance in general - read some papers explaining that on 200MHz
PentiumPro systems the bottleneck was RAM bandwidth!)
The buffer-cache reads timing (Mark was referring to) is sometimes a
good index:
Timing buffer-cache reads: 128 MB in 0.80 seconds =160.00 MB/sec
Timing buffered disk reads: 64 MB in 1.79 seconds = 35.75 MB/sec
(Athlon 700 slot-A system, not idle. I've seen 165+. Also note disk b/w.)
Timing buffer-cache reads: 128 MB in 0.58 seconds =220.69 MB/sec
Timing buffered disk reads: 64 MB in 1.85 seconds = 34.59 MB/sec
(Athlon 1333 DDR system, not idle. On initlevel 1 I've seen 245+)
Now, 220 vs 109 is a lot. But even 160 vs 109 is. Even the slowest system
I've seen (Duron 700MHz) was above 140. I bet your old Althon system
was capable of at least 145/150. So you downgraded, that's all.
rule 1: don't buy MB with integrated (shared mem) video chips;
rule 2: use DDR capable MB and CPU, @ 133MHz (or 266 on ad material);
rule 3: use SCSI if you can afford it. But ATA is fine on low-end.
>
> > doesn't the i815 have integrated video,
>
> yes
>
> > and if so, are you using it,
>
> yes
>
> > and if so, do you have the optional "display cache"?
>
> couldn't find any BIOS setting for that. where should it be ?
>
> > without the DC, video steals quite a lot of dram bandwidth...
>
> the machine is (should be) a NFS server, I couldn't care less about the
> video
it manages to steal mem b/w anyway.
Don't use that MB (and Celerons) for servers.
.TM.
--
____/ ____/ /
/ / / Marco Colombo
___/ ___ / / Technical Manager
/ / / ESI s.r.l.
_____/ _____/ _/ Colombo@ESI.it
^ permalink raw reply [flat|nested] 15+ messages in thread
* disk transfer speed problem
@ 2002-02-27 20:49 lonely wolf
2002-02-28 2:06 ` Tim Moore
2002-02-28 22:48 ` Stephen Samuel
0 siblings, 2 replies; 15+ messages in thread
From: lonely wolf @ 2002-02-27 20:49 UTC (permalink / raw)
To: linux-kernel
initial data: one brand new intel bonham motherboard (i815) , 900 MHz
celeron processor
2 new 80 GB Seagate Barracuda used as raid1, 7200 RPM,ext3 filesystem,
RedHat 2.4.9-21 kernel from RH 7.2 updates.
settings (identical for hdc and hdd), as detected by the kernel (except
for I/O support = 3):
#hdparm /dev/hdc
multcount = 16 (on)
I/O support = 3 (32-bit w/sync)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 23989/16/63, sectors = 156301488, start = 0
#hdparm -i /dev/hdc
Model=ST380021A, FwRev=3.05, SerialNo=3HV080KH
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=156301488
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
#dmesg|grep UDMA
hda: 14668290 sectors (7510 MB) w/418KiB Cache, CHS=913/255/63, UDMA(66)
(this is the boot disk)
hdc: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
UDMA(100
hdd: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
UDMA(100)
The IDE cables have 80 wires. I am not sure that they are ATA100, but I
think they do.
The problem:
#/sbin/hdparm -tTf /dev/hdd
/dev/hdd:
Timing buffer-cache reads: 128 MB in 1.17 seconds =109.40 MB/sec
Timing buffered disk reads: 64 MB in 3.24 seconds = 19.75 MB/sec
Even after issuing raidstop and umounting the partition placed on the
RAID, the speed does not exceed 22 MB/sec.
With hdparm -c1 performances are almost identical
Tests were performed in initlevel 3, with NFS started, but no one but
root using the machine (from console). In production, the speed is also
very, very slow
Any idea what might limit the transfer speed ?
--
Manuel Wolfshant linux registered user #131416
network administrator NoBug Consulting Romania
Beware the fury of a patient man.
^ permalink raw reply [flat|nested] 15+ messages in thread* Re: disk transfer speed problem
2002-02-27 20:49 lonely wolf
@ 2002-02-28 2:06 ` Tim Moore
2002-02-28 4:51 ` Andre Hedrick
2002-02-28 22:48 ` Stephen Samuel
1 sibling, 1 reply; 15+ messages in thread
From: Tim Moore @ 2002-02-28 2:06 UTC (permalink / raw)
To: linux-kernel
> #dmesg|grep UDMA
> hda: 14668290 sectors (7510 MB) w/418KiB Cache, CHS=913/255/63, UDMA(66)
> (this is the boot disk)
>
> hdc: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
> UDMA(100
> hdd: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
> UDMA(100)
Please post dmesg, /proc/pci and chipset info from /proc/ide. Also,
what is result from 'hdparm -tT /dev/hda'?
rgds,
tim.
--
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-28 2:06 ` Tim Moore
@ 2002-02-28 4:51 ` Andre Hedrick
2002-02-28 11:56 ` Dave Jones
2002-02-28 12:27 ` Martin Dalecki
0 siblings, 2 replies; 15+ messages in thread
From: Andre Hedrick @ 2002-02-28 4:51 UTC (permalink / raw)
To: Tim Moore; +Cc: linux-kernel
What is more useful is the cat /proc/ide/ide0/config !!!
>From that I can tell you what is going on completely about the system.
Oh and for those not reading this email, it is a side note on why the ide
proc-pci interface had best be left alone and in tact.
Cheers,
Andre Hedrick
Linux Disk Certification Project Linux ATA Development
On Wed, 27 Feb 2002, Tim Moore wrote:
> > #dmesg|grep UDMA
> > hda: 14668290 sectors (7510 MB) w/418KiB Cache, CHS=913/255/63, UDMA(66)
> > (this is the boot disk)
> >
> > hdc: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
> > UDMA(100
> > hdd: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
> > UDMA(100)
>
> Please post dmesg, /proc/pci and chipset info from /proc/ide. Also,
> what is result from 'hdparm -tT /dev/hda'?
>
> rgds,
> tim.
> --
> -
> 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/
>
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-28 4:51 ` Andre Hedrick
@ 2002-02-28 11:56 ` Dave Jones
2002-02-28 17:14 ` Andre Hedrick
2002-02-28 12:27 ` Martin Dalecki
1 sibling, 1 reply; 15+ messages in thread
From: Dave Jones @ 2002-02-28 11:56 UTC (permalink / raw)
To: Andre Hedrick; +Cc: Tim Moore, linux-kernel
On Wed, Feb 27, 2002 at 08:51:44PM -0800, Andre Hedrick wrote:
>
> What is more useful is the cat /proc/ide/ide0/config !!!
> Oh and for those not reading this email, it is a side note on why the ide
> proc-pci interface had best be left alone and in tact.
And lspci -xxx is insufficent because?
--
| Dave Jones. http://www.codemonkey.org.uk
| SuSE Labs
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-28 11:56 ` Dave Jones
@ 2002-02-28 17:14 ` Andre Hedrick
0 siblings, 0 replies; 15+ messages in thread
From: Andre Hedrick @ 2002-02-28 17:14 UTC (permalink / raw)
To: Dave Jones; +Cc: Tim Moore, linux-kernel
On Thu, 28 Feb 2002, Dave Jones wrote:
> On Wed, Feb 27, 2002 at 08:51:44PM -0800, Andre Hedrick wrote:
> >
> > What is more useful is the cat /proc/ide/ide0/config !!!
> > Oh and for those not reading this email, it is a side note on why the ide
> > proc-pci interface had best be left alone and in tact.
>
> And lspci -xxx is insufficent because?
You and I know it is the functional equivalent; however, in development of
driver it is more practical and faster to issue
echo PXX:YY > config
setpci XX.YY.Z blah blah ...
So if there is a real issue with its presense, apply a uuid test for
access. Everything else has it so this should too.
Cheers,
Andre Hedrick
Linux Disk Certification Project Linux ATA Development
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-28 4:51 ` Andre Hedrick
2002-02-28 11:56 ` Dave Jones
@ 2002-02-28 12:27 ` Martin Dalecki
1 sibling, 0 replies; 15+ messages in thread
From: Martin Dalecki @ 2002-02-28 12:27 UTC (permalink / raw)
To: Andre Hedrick; +Cc: Tim Moore, linux-kernel
Andre Hedrick wrote:
> What is more useful is the cat /proc/ide/ide0/config !!!
>
>>From that I can tell you what is going on completely about the system.
>
> Oh and for those not reading this email, it is a side note on why the ide
> proc-pci interface had best be left alone and in tact.
Module the fact fact lspci could be easly extendid to provide the
same information from user space... wait... we are on unix and we have
manual pages:
-x Show hexadecimal dump of first 64 bytes of the PCI
configuration space (the standard header). Useful
for debugging of drivers and lspci itself.
-xxx Show hexadecimal dump of whole PCI configuration
space. Available only for root as several PCI
devices crash when you try to read undefined por-
tions of the config space (this behaviour probably
doesn't violate the PCI standard, but it's at least
very stupid).
As you can see lspci -x -xxx ALREADY DOES THIS!
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: disk transfer speed problem
2002-02-27 20:49 lonely wolf
2002-02-28 2:06 ` Tim Moore
@ 2002-02-28 22:48 ` Stephen Samuel
2002-02-28 23:58 ` lonely wolf
1 sibling, 1 reply; 15+ messages in thread
From: Stephen Samuel @ 2002-02-28 22:48 UTC (permalink / raw)
To: lonely wolf; +Cc: linux-kernel
Picking nits, perhaps -- but after giving us data on hdc,
you did the test on hdd. I realize you said that it is identical
for both, but if you missed something critical, we'll never see it.
lonely wolf wrote:
> initial data: one brand new intel bonham motherboard (i815) , 900 MHz
> celeron processor
> 2 new 80 GB Seagate Barracuda used as raid1, 7200 RPM,ext3 filesystem,
> RedHat 2.4.9-21 kernel from RH 7.2 updates.
>
> settings (identical for hdc and hdd), as detected by the kernel (except
> for I/O support = 3):
> #hdparm /dev/hdc
> multcount = 16 (on)
> I/O support = 3 (32-bit w/sync)
> unmaskirq = 0 (off)
> using_dma = 1 (on)
> keepsettings = 0 (off)
> nowerr = 0 (off)
> readonly = 0 (off)
> readahead = 8 (on)
> geometry = 23989/16/63, sectors = 156301488, start = 0
>
> #hdparm -i /dev/hdc
>
> Model=ST380021A, FwRev=3.05, SerialNo=3HV080KH
> Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
> RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
> BuffType=unknown, BuffSize=2048kB, MaxMultSect=16, MultSect=16
> CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=156301488
> IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
> PIO modes: pio0 pio1 pio2 pio3 pio4
> DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
>
> #dmesg|grep UDMA
> hda: 14668290 sectors (7510 MB) w/418KiB Cache, CHS=913/255/63, UDMA(66)
> (this is the boot disk)
>
> hdc: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
> UDMA(100
> hdd: 156301488 sectors (80026 MB) w/2048KiB Cache, CHS=155061/16/63,
> UDMA(100)
>
> The IDE cables have 80 wires. I am not sure that they are ATA100, but I
> think they do.
>
> The problem:
> #/sbin/hdparm -tTf /dev/hdd
>
> /dev/hdd:
> Timing buffer-cache reads: 128 MB in 1.17 seconds =109.40 MB/sec
> Timing buffered disk reads: 64 MB in 3.24 seconds = 19.75 MB/sec
>
>
> Even after issuing raidstop and umounting the partition placed on the
> RAID, the speed does not exceed 22 MB/sec.
>
> With hdparm -c1 performances are almost identical
>
> Tests were performed in initlevel 3, with NFS started, but no one but
> root using the machine (from console). In production, the speed is also
> very, very slow
>
> Any idea what might limit the transfer speed ?
>
>
> --
> Manuel Wolfshant linux registered user #131416
> network administrator NoBug Consulting Romania
> Beware the fury of a patient man.
>
>
>
>
> -
> 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/
>
--
Stephen Samuel +1(604)876-0426 samuel@bcgreen.com
http://www.bcgreen.com/~samuel/
Powerful committed communication, reaching through fear, uncertainty and
doubt to touch the jewel within each person and bring it to life.
^ permalink raw reply [flat|nested] 15+ messages in thread* Re: disk transfer speed problem
2002-02-28 22:48 ` Stephen Samuel
@ 2002-02-28 23:58 ` lonely wolf
0 siblings, 0 replies; 15+ messages in thread
From: lonely wolf @ 2002-02-28 23:58 UTC (permalink / raw)
Cc: linux-kernel
On Thu, 28 Feb 2002, Stephen Samuel wrote:
> Picking nits, perhaps -- but after giving us data on hdc,
> you did the test on hdd. I realize you said that it is identical
> for both, but if you missed something critical, we'll never see it.
they are 100% identical. except for the fact that one is
master and the other slave on the same cable.
I did not include the results for both in order to shorten the message.
here we go, just in case:
[root@bugmaster /root]# /sbin/hdparm -tT /dev/hdc /dev/hdd;/sbin/hdparm
/dev/hdc /dev/hdd;/sbin/hdparm -i /dev/hdc /dev/hdd
/dev/hdc:
Timing buffer-cache reads: 128 MB in 1.17 seconds =109.40 MB/sec
Timing buffered disk reads: 64 MB in 3.01 seconds = 21.26 MB/sec
/dev/hdd:
Timing buffer-cache reads: 128 MB in 1.09 seconds =117.43 MB/sec
Timing buffered disk reads: 64 MB in 2.96 seconds = 21.62 MB/sec
/dev/hdc:
multcount = 16 (on)
I/O support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 23989/16/63, sectors = 156301488, start = 0
/dev/hdd:
multcount = 16 (on)
I/O support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 23989/16/63, sectors = 156301488, start = 0
/dev/hdc:
Model=ST380021A, FwRev=3.05, SerialNo=3HV080KH
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=156301488
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
/dev/hdd:
Model=ST380021A, FwRev=3.05, SerialNo=3HV07SDN
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=156301488
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
Mr Davidovac Zoran mentioned normal transfer speeds with kernel 2.4.17.
Should I try ? Are there significant differences between 2.4.9-21 (or -31)
and 2.4.17 in the IDE drivers ?
-
This is Linux Country. On a quiet night, you can hear
Windows NT reboot!
^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2002-03-01 11:06 UTC | newest]
Thread overview: 15+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <Pine.LNX.4.33.0202271701360.13103-100000@coffee.psychology.mcmaster.ca>
2002-02-27 22:52 ` disk transfer speed problem lonely wolf
2002-02-27 23:46 ` Alan Cox
2002-02-28 9:20 ` Davidovac Zoran
2002-02-28 14:49 ` lonely wolf
2002-03-01 2:25 ` lonely wolf
2002-03-01 10:59 ` Andre Hedrick
2002-03-01 10:51 ` Marco Colombo
2002-02-27 20:49 lonely wolf
2002-02-28 2:06 ` Tim Moore
2002-02-28 4:51 ` Andre Hedrick
2002-02-28 11:56 ` Dave Jones
2002-02-28 17:14 ` Andre Hedrick
2002-02-28 12:27 ` Martin Dalecki
2002-02-28 22:48 ` Stephen Samuel
2002-02-28 23:58 ` lonely wolf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox