| View previous topic :: View next topic |
| Author |
Message |
raulpober
n00b
Joined: 26 Oct 2006
Posts: 36
|
 Posted: Thu Jul 12, 2007 6:23 pm Post subject: Very inconsistent results from hdparm |
 |
|
I'm working on speeding up the disk access on an older machine, and I'm finding some significant inconsistencies in the output from hdparm -tT:
| Code: |
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 532 MB in 2.01 seconds = 264.97 MB/sec
Timing buffered disk reads: 88 MB in 3.01 seconds = 29.24 MB/sec
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 556 MB in 2.00 seconds = 277.33 MB/sec
Timing buffered disk reads: 14 MB in 3.06 seconds = 4.58 MB/sec
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 552 MB in 2.00 seconds = 275.49 MB/sec
Timing buffered disk reads: 2 MB in 9.96 seconds = 205.65 kB/sec
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 548 MB in 2.01 seconds = 273.19 MB/sec
Timing buffered disk reads: 2 MB in 4.09 seconds = 500.66 kB/sec
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 548 MB in 2.00 seconds = 273.90 MB/sec
Timing buffered disk reads: 2 MB in 11.59 seconds = 176.68 kB/sec
|
hdparm -i gives:
| Code: |
# hdparm -i /dev/hda
/dev/hda:
Model=SAMSUNG SV8004H, FwRev=QR100-12, SerialNo=0422J1FT907935
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=38997, SectSize=619, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=156368016
IORDY=yes, 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
AdvancedPM=no WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 1: ATA/ATAPI-1,2,3,4,5,6
* signifies the current active mode
|
and the settings are (fron Gentoo handbook):
| Code: |
# hdparm -v /dev/hda
/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 64 (on)
geometry = 65535/16/63, sectors = 156368016, start = 0
|
Not that familiar with hard drive tuning, but I would have expected the performance tests to be more consistent. Any ideas on what might be causing the buffered reads to slow down so much after a few tests? I do notice that disk access is very slow on the machine. |
|
| Back to top |
|
 |
eccerr0r
Watchman
Joined: 01 Jul 2004
Posts: 9882
Location: almost Mile High in the USA
|
| Posted: Thu Jul 12, 2007 8:42 pm Post subject: |
|
|
Sometimes it's because the disk is dying?
Check dmesg to see if there are any errors reported. Also try installing smartmontools to see the condition of the disk, if it supports SMART.
_________________
Intel Core i7 2700K/Radeon R7 250/24GB DDR3/256GB SSD
What am I supposed watching? |
|
| Back to top |
|
|
raulpober
n00b
Joined: 26 Oct 2006
Posts: 36
|
| Posted: Thu Jul 12, 2007 9:52 pm Post subject: |
|
|
I don't see any error messages in dmesg
| Code: |
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
VP_IDE: IDE controller at PCI slot 0000:00:11.1
ACPI: Unable to derive IRQ for device 0000:00:11.1
ACPI: PCI Interrupt 0000:00:11.1[A]: no GSI
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
VP_IDE: VIA vt8235 (rev 00) IDE UDMA133 controller on pci0000:00:11.1
ide0: BM-DMA at 0xfc00-0xfc07, BIOS settings: hda:DMA, hdb:pio
ide1: BM-DMA at 0xfc08-0xfc0f, BIOS settings: hdc:DMA, hdd:DMA
Probing IDE interface ide0...
hda: SAMSUNG SV8004H, ATA DISK drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
|
I wish I had some records of speed tests in the past to compare, but I don't. I'll check into the SMART monitoring. |
|
| Back to top |
|
|
eccerr0r
Watchman
Joined: 01 Jul 2004
Posts: 9882
Location: almost Mile High in the USA
|
| Posted: Tue Jul 24, 2007 12:38 am Post subject: |
|
|
Make sure about the obvious stuff like no updatedb running in the background...
as for smartmontools, smartmon -a /dev/hda and see if there's any failing states there.
You should be getting consistent 20+MB/sec even with the "older" machine, the numbers in the KB/sec range is really wrong. My oldest in-service machine is a P5-166MMX and can still get around 10MB/sec consistently through hdparm.
As another silly thing: how much RAM is in the machine and are you swapping when trying to run hdparm?
_________________
Intel Core i7 2700K/Radeon R7 250/24GB DDR3/256GB SSD
What am I supposed watching? |
|
| Back to top |
|
|
raulpober
n00b
Joined: 26 Oct 2006
Posts: 36
|
| Posted: Mon Jul 30, 2007 8:21 am Post subject: |
|
|
Yep. Turned out to be a silly thing. I only had 512 MB of ram and was runing climate models through boinc. Although I had suspended boinc, the climate model takes up a lot of ram, so I think the swapping idea is most likely. After adding another GB, I now get:
| Code: |
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 316 MB in 2.01 seconds = 157.47 MB/sec
Timing buffered disk reads: 96 MB in 3.01 seconds = 31.85 MB/sec
# hdparm -tT /dev/hda
/dev/hda:
Timing cached reads: 320 MB in 2.01 seconds = 159.16 MB/sec
Timing buffered disk reads: 96 MB in 3.01 seconds = 31.87 MB/sec
hellken paul # smartmon -a /dev/hda
So, all seems to be in order. Thanks for the comments! |
|
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
