Re: Slow HDD speed on Dell E6400
TomC!E! BodEC!r wrote: There must be some funny tricks on those other OS's.But it doesn't matter (I will investigate myself). Now I know more about those random generators and tests for real are ok. Untar of src.tar.gz shows about 9MB/s in iostat(8) and dd ports.tar.gz to some file shows about 22MB/s. Thanks all for their tips and sorry for some of my stupid ideas ;-) Also, untarring lots of small files wont test your I/O in the same way as a dd(1) test, since it will test how often can I make atomic writes to my disk rather than how much data can I move.
Re: Slow HDD speed on Dell E6400
I know.That's why I tested in dd and iostat too. 2009/9/29 Janne Johansson j...@it.su.se: TomC!E! BodEC!r wrote: There must be some funny tricks on those other OS's.But it doesn't matter (I will investigate myself). Now I know more about those random generators and tests for real are ok. Untar of src.tar.gz shows about 9MB/s in iostat(8) and dd ports.tar.gz to some file shows about 22MB/s. Thanks all for their tips and sorry for some of my stupid ideas ;-) Also, untarring lots of small files wont test your I/O in the same way as a dd(1) test, since it will test how often can I make atomic writes to my disk rather than how much data can I move. -- http://www.openbsd.org/lyrics.html
Re: Slow HDD speed on Dell E6400
Here is my e6500: $ sudo dd if=/dev/rsd0c of=/dev/null bs=1m count=500 500+0 records in 500+0 records out 524288000 bytes transferred in 4.631 secs (113202161 bytes/sec) On Tue, Sep 29, 2009 at 09:44:53AM +0200, Tom Bodr wrote: I know.That's why I tested in dd and iostat too. 2009/9/29 Janne Johansson j...@it.su.se: TomC!E! BodEC!r wrote: There must be some funny tricks on those other OS's.But it doesn't matter (I will investigate myself). Now I know more about those random generators and tests for real are ok. Untar of src.tar.gz shows about 9MB/s in iostat(8) and dd ports.tar.gz to some file shows about 22MB/s. Thanks all for their tips and sorry for some of my stupid ideas ;-) Also, untarring lots of small files wont test your I/O in the same way as a dd(1) test, since it will test how often can I make atomic writes to my disk rather than how much data can I move. -- http://www.openbsd.org/lyrics.html
Slow HDD speed on Dell E6400
Hi all, when I try dd command I will get similar numbers : $ dd if=/dev/urandom of=test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 6.798 secs (154233 bytes/sec) $ On my old desktop with Ubuntu I have about 1,7MB/s,my friends with Linux have from about 3 to 8MB/s,my friend with OpenBSD on server has about 50MB/s(of course totally different HW) and what's worse on same laptop I have about 33MB/s with OpenSolaris.Of course that this is not some magic test of speed,but it is visible how slow is it even when working with apps.I think that there is some problem but I can't find where.I tried to use SoftDependencies,turn off atime,I played with namei cache size and all without success.I'm reading trough man pages for drivers and so on,but can't find anything obvious. (same results are on amd64). Problem of chipset,disk,some option? $ apm Battery state: high, 100% remaining, unknown life estimate A/C adapter state: connected Performance adjustment mode: cool running (800 MHz) $ $ sudo atactl wd0 identify Model: Hitachi HTS723280L9A362, Rev: FC1OC30F, Serial #: S/N Device type: ATA, fixed Cylinders: 16383, heads: 16, sec/track: 63, total sectors: 156301488 Device capabilities: IORDY operation IORDY disabling Device supports the following standards: ATA-2 ATA-3 ATA-4 ATA-5 ATA-6 ATA-7 ATA-8 Master password revision code 0xfffe Device supports the following command sets: NOP command READ BUFFER command WRITE BUFFER command Host Protected Area feature set Read look-ahead Write cache Power Management feature set Security Mode feature set SMART feature set Flush Cache Ext command Flush Cache command Device Configuration Overlay feature set 48bit address feature set Automatic Acoustic Management feature set Set Max security extension commands Set Features subcommand required Power-up in standby feature set Advanced Power Management feature set DOWNLOAD MICROCODE command IDLE IMMEDIATE with UNLOAD FEATURE SMART self-test SMART error logging Device has enabled the following command sets/features: NOP command READ BUFFER command WRITE BUFFER command Host Protected Area feature set Read look-ahead Write cache Power Management feature set SMART feature set Flush Cache Ext command Flush Cache command Device Configuration Overlay feature set 48bit address feature set Automatic Acoustic Management feature set Set Features subcommand required Advanced Power Management feature set DOWNLOAD MICROCODE command $ $ cat /etc/fstab /dev/wd0a / ffs rw,softdep 1 1 /dev/wd0l /home ffs rw,nodev,nosuid,softdep 1 2 /dev/wd0d /tmp ffs rw,nodev,nosuid,softdep 1 2 /dev/wd0f /usr ffs rw,nodev,softdep 1 2 /dev/wd0h /usr/X11R6 ffs rw,nodev,softdep 1 2 /dev/wd0g /usr/local ffs rw,nodev,softdep 1 2 /dev/wd0j /usr/obj ffs rw,nodev,nosuid,softdep 1 2 /dev/wd0k /usr/src ffs rw,nodev,nosuid,softdep 1 2 /dev/wd0e /var ffs rw,nodev,nosuid,softdep 1 2 $ $ df -h Filesystem SizeUsed Avail Capacity Mounted on /dev/wd0a 254M 43.0M198M18%/ /dev/wd0l 9.6G262M8.9G 3%/home /dev/wd0d 1008M8.0K958M 0%/tmp /dev/wd0f 3.4G424M2.8G13%/usr /dev/wd0h 508M166M317M34%/usr/X11R6 /dev/wd0g 2.9G158M2.6G 6%/usr/local /dev/wd0j 1.9G2.0K1.8G 0%/usr/obj /dev/wd0k 1.9G2.0K1.8G 0%/usr/src /dev/wd0e 254M8.2M233M 3%/var $ $ swapctl -lk Device 1K-blocks UsedAvail Capacity Priority swap_device 40162040162 0%0 $ $ dmesg OpenBSD 4.6-current (GENERIC.MP) #205: Thu Sep 24 10:57:47 MDT 2009 t...@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC.MP cpu0: Intel(R) Core(TM)2 Duo CPU P8600 @ 2.40GHz (GenuineIntel 686-class) 2.40 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,SBF,SSE3,MWAIT,DS-CPL,VMX,SMX,EST,TM2,CX16,xTPR real mem = 3707650048 (3535MB) avail mem = 3608420352 (3441MB) mainbus0 at root bios0 at mainbus0: AT/286+ BIOS, date 02/13/09, BIOS32 rev. 0 @ 0xffa10, SMBIOS rev. 2.4 @ 0xf6590 (57 entries) bios0: vendor Dell Inc. version A12 date 02/13/2009 bios0: Dell Inc. Latitude E6400 acpi0 at bios0: rev 2 acpi0: tables DSDT FACP HPET DMAR APIC ASF! MCFG TCPA SLIC SSDT acpi0: wakeup devices PCI0(S4) PCIE(S4) USB1(S3) USB2(S3) USB3(S3) USB4(S3) USB5(S3) USB6(S3) EHC2(S3) EHCI(S3) AZAL(S3) RP01(S4) RP02(S4) RP03(S4) RP04(S3) RP05(S3) RP06(S5) LID_(S3) PBTN(S4) acpitimer0 at acpi0: 3579545 Hz, 24 bits acpihpet0 at acpi0: 14318179 Hz acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0
Re: Slow HDD speed on Dell E6400
2009/9/28 TomC!E! BodEC!r tomas.bod...@gmail.com: when I try dd command I will get similar numbers : $ dd if=/dev/urandom of=test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 6.798 secs (154233 bytes/sec) On my old desktop with Ubuntu I have about 1,7MB/s,my friends with Linux have from about 3 to 8MB/s,my friend with OpenBSD on server has about 50MB/s(of course totally different HW) and what's worse on same laptop I have about 33MB/s with OpenSolaris.Of course that this is not some magic test of speed,but it is visible how slow is it even when working with apps.I think that there is some problem but I can't find where.I tried to use SoftDependencies,turn off atime,I played with namei cache size and all without success.I'm reading trough man pages for drivers and so on,but can't find anything obvious. (same results are on amd64). Problem of chipset,disk,some option? Are you testing the speed of urandom or your HD? If the latter, you might want to use something like /dev/zero instead.
Re: Slow HDD speed on Dell E6400
2009/9/28 TomC!E! BodEC!r tomas.bod...@gmail.com: when I try dd command I will get similar numbers : $ dd if=/dev/urandom of=test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 6.798 secs (154233 bytes/sec) $ On my old desktop with Ubuntu I have about 1,7MB/s,my friends with Linux have from about 3 to 8MB/s,my friend with OpenBSD on server has about 50MB/s(of course totally different HW) and what's worse on same laptop I have about 33MB/s with OpenSolaris.Of course that this is not some magic test of speed,but it is visible how slow is it even when working with apps.I think that there is some problem but I can't find where.I tried to use SoftDependencies,turn off atime,I played with namei cache size and all without success.I'm reading trough man pages for drivers and so on,but can't find anything obvious. (same results are on amd64). Problem of chipset,disk,some option? How fast can you read from /dev/urandom?
Re: Slow HDD speed on Dell E6400
Thanks to all for points.Now I'm dived in man pages :-) For disk there is a option for AHCI mode,but not possible on my laptop. I have Win in dual boot and they don't like AHCI heh. For urandom I'm reading man pages around it on Linux and OpenBSD to try find difference. On Mon, Sep 28, 2009 at 4:15 PM, Daniel Melameth dan...@melameth.com wrote: 2009/9/28 TomC!E! BodEC!r tomas.bod...@gmail.com: when I try dd command I will get similar numbers : $ dd if=/dev/urandom of=test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 6.798 secs (154233 bytes/sec) On my old desktop with Ubuntu I have about 1,7MB/s,my friends with Linux have from about 3 to 8MB/s,my friend with OpenBSD on server has about 50MB/s(of course totally different HW) and what's worse on same laptop I have about 33MB/s with OpenSolaris.Of course that this is not some magic test of speed,but it is visible how slow is it even when working with apps.I think that there is some problem but I can't find where.I tried to use SoftDependencies,turn off atime,I played with namei cache size and all without success.I'm reading trough man pages for drivers and so on,but can't find anything obvious. (same results are on amd64). Problem of chipset,disk,some option? Are you testing the speed of urandom or your HD? B If the latter, you might want to use something like /dev/zero instead. -- http://www.openbsd.org/lyrics.html
Re: Slow HDD speed on Dell E6400
On Mon, Sep 28, 2009 at 04:35:59PM +0200, TomC!E! BodEC!r wrote: On Mon, Sep 28, 2009 at 4:15 PM, Daniel Melameth dan...@melameth.com wrote: 2009/9/28 TomC!E! BodEC!r tomas.bod...@gmail.com: when I try dd command I will get similar numbers : $ dd if=/dev/urandom of=test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 6.798 secs (154233 bytes/sec) On my old desktop with Ubuntu I have about 1,7MB/s,my friends with Linux have from about 3 to 8MB/s, [SNIP: more disk-performance related guesses -- Joachim]. Are you testing the speed of urandom or your HD? If the latter, you might want to use something like /dev/zero instead. Thanks to all for points.Now I'm dived in man pages :-) For disk there is a option for AHCI mode,but not possible on my laptop. I have Win in dual boot and they don't like AHCI heh. For urandom I'm reading man pages around it on Linux and OpenBSD to try find difference. Huh? There is no need to read man pages, just check $ dd if=/dev/urandom of=/dev/null bs=1k count=1024 for a reasonable upper bound on the performance of dd reading from /dev/urandom. You may find that it is very close to the above numbers, i.e. the disk is not the bottleneck. You can then repeat with /dev/zero, as suggested. If you are worried about the predictable pattern, use /dev/arandom, which is a lot faster than /dev/urandom - you don't need cryptographically secure random numbers, after all. On my machine, for instance, $ dd if=/dev/urandom of=/dev/null bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 9.143 secs (114683 bytes/sec) $ dd if=/dev/zero of=/dev/null bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 0.002 secs (379094722 bytes/sec) $ dd if=/dev/arandom of=/dev/null bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 0.040 secs (25746458 bytes/sec) $ dd if=/dev/arandom of=$HOME/test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 0.547 secs (1915326 bytes/sec) In other words, urandom is noticeably slower than the disk. On a higher level, you aren't really clear what you are trying to measure, but if it's disk performance there are a lot of factors you haven't considered. Repeating your experiment with a larger count and/or block size may be instructive (for instance, I saw a 25% loss in performance going to count=16384.) In fact, I'm pretty sure that someone with a strong Linux background could persuade that OS to cache the complete write in memory... Joachim
Re: Slow HDD speed on Dell E6400
There must be some funny tricks on those other OS's.But it doesn't matter (I will investigate myself). Now I know more about those random generators and tests for real are ok. Untar of src.tar.gz shows about 9MB/s in iostat(8) and dd ports.tar.gz to some file shows about 22MB/s. Thanks all for their tips and sorry for some of my stupid ideas ;-) Small donation sent and 3 T-shirts purchased. On Mon, Sep 28, 2009 at 5:25 PM, Joachim Schipper joac...@joachimschipper.nl wrote: On Mon, Sep 28, 2009 at 04:35:59PM +0200, TomC!E! BodEC!r wrote: On Mon, Sep 28, 2009 at 4:15 PM, Daniel Melameth dan...@melameth.com wrote: 2009/9/28 TomC!E! BodEC!r tomas.bod...@gmail.com: when I try dd command I will get similar numbers : $ dd if=/dev/urandom of=test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 6.798 secs (154233 bytes/sec) On my old desktop with Ubuntu I have about 1,7MB/s,my friends with Linux have from about 3 to 8MB/s, [SNIP: more disk-performance related guesses -- Joachim]. Are you testing the speed of urandom or your HD? If the latter, you might want to use something like /dev/zero instead. Thanks to all for points.Now I'm dived in man pages :-) For disk there is a option for AHCI mode,but not possible on my laptop. I have Win in dual boot and they don't like AHCI heh. For urandom I'm reading man pages around it on Linux and OpenBSD to try find difference. Huh? There is no need to read man pages, just check $ dd if=/dev/urandom of=/dev/null bs=1k count=1024 for a reasonable upper bound on the performance of dd reading from /dev/urandom. You may find that it is very close to the above numbers, i.e. the disk is not the bottleneck. You can then repeat with /dev/zero, as suggested. If you are worried about the predictable pattern, use /dev/arandom, which is a lot faster than /dev/urandom - you don't need cryptographically secure random numbers, after all. On my machine, for instance, $ dd if=/dev/urandom of=/dev/null bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 9.143 secs (114683 bytes/sec) $ dd if=/dev/zero of=/dev/null bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 0.002 secs (379094722 bytes/sec) $ dd if=/dev/arandom of=/dev/null bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 0.040 secs (25746458 bytes/sec) $ dd if=/dev/arandom of=$HOME/test bs=1k count=1024 1024+0 records in 1024+0 records out 1048576 bytes transferred in 0.547 secs (1915326 bytes/sec) In other words, urandom is noticeably slower than the disk. On a higher level, you aren't really clear what you are trying to measure, but if it's disk performance there are a lot of factors you haven't considered. Repeating your experiment with a larger count and/or block size may be instructive (for instance, I saw a 25% loss in performance going to count=16384.) In fact, I'm pretty sure that someone with a strong Linux background could persuade that OS to cache the complete write in memory... B B B B B B B B Joachim -- http://www.openbsd.org/lyrics.html
