Re: Very slow writing to SATA disk
On Sat, Oct 29, 2005 at 01:00:36AM +0200, S?ren Schmidt wrote: On 29/10/2005, at 0:41, Mikhail Teterin wrote: Look in smartmontools I provided patches for that, its not rocket science you know... This attitude -- on top of the API change itself -- is not really encouraging for ISVs, you know :-) Sigh, ataidle is a hack and the author had no intention to listen back when, so I dont feel teribly sorry about it you know. Spinning down disks needs to be done at the driver level so ATA knows what state the disk is in etc... I'm the author of ataidle. I know it is a hack, and I agree that power management should be done at the driver level. However, I have released what worked for me in the hope that other people find it useful - but yes, there is always the possibility that data will be lost, since it does bypass the driver. Unfortunately I've been very busy recently and so have until now been unable to look at FreeBSD 6.0 and the new ATA driver. I have however updated ataidle to cope with the new API and it can be downloaded from http://www.cran.org.uk/bruce/software/ataidle-0.9.tar.gz . -- Bruce Cran ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
If this drive doesn't support tagged-queueing, is the write cache disabled? I get that sort of performance from a (PATA) disk with the cache disabled (hw.ata.wc=0 in loader.conf) No, just checked -- the hw.ata.wc is set to 1. Is there anything else to look at? According to smartctl, the drive runs at 56C during the copying. Its idle temperature seems to be 54C. I'd double-check that (eg with a finger). If the drive really is running at 56°C, it won't last very long. It sure feels hot to the touch, but nothing is burning, of course (freshly poured tee is near 100C and never ignites the paper cup). Do you think, the high temperature explains the poor write performance? The drive still reads at tens of Mb per second... According to http://www.spacecentersystems.com/catalog/product_info.php/products_id/293088 this model's 'Ambient Temperature' spec is 5 to 55C... I'd like to be able to turn the drive off between backups, but sysutils/ataidle is broken on 6.0 (Soren?). Thanks! -mi ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
Mikhail Teterin wrote: idle temperature seems to be 54C. According to smartctl, the drive runs at 56C during the copying. Its I'd double-check that (eg with a finger). If the drive really is running at 56°C, it won't last very long. It sure feels hot to the touch, but nothing is burning, of course (freshly poured tee is near 100C and never ignites the paper cup). Do you think, the high temperature explains the poor write performance? The drive still reads at tens of Mb per second... According to http://www.spacecentersystems.com/catalog/product_info.php/products_id/293088 this model's 'Ambient Temperature' spec is 5 to 55C... Generally, the temperature is not directly affecting the performance till the mechanical problem occurred due to the temperature. A few types of hard drives are manufactured with higher temperature spec., however, those drives are usually broken in 3-6 months (continuously run). Otherwise, the higher temperature is due to some mechanical problem which may slow down the performance. In your case, it seems to me that your drive temperature is at or above the upper bound that could indicate a mechanical problem. Check to see if the reading speed of this drive is normal, then, this won't be the case for now. But keep in mind, if this drive always operates at this temperature, it will not last long. The better temperature for electronic components is below 39C, and not more than 45C~50C. So, 56C can gradually damage the mechanical as well as the electronic components. -Jin ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
Indeed, 55C is way to high for 24/7 usage, and it might be that the drive is choking on it and barely is able to compensate.. The reads are pretty quick... I'd like to be able to spin it down, but ataidle is broken :-( What does SMART say ? any unusual like high correction rates or anything ? % atacontrol ad8 Protocol Serial ATA v1.0 Serial ATA II device model HDS725050KLA360 serial number KRVN02ZAG1SV7C firmware revision K2AOA11A cylinders 16383 heads 16 sectors/track 63 lba supported 268435455 sectors lba48 supported 976773168 sectors dma supported overlap not supported Feature Support EnableValue Vendor write cacheyes yes read ahead yes yes Native Command Queuing (NCQ) yes - 31/0x1F Tagged Command Queuing (TCQ) no no 31/0x1F SMART yes yes microcode download yes yes security yes no power management yes yes advanced power management yes no 0/0x00 automatic acoustic management yes yes 128/0x80128/0x80 % smartctl -a /dev/ad8 === START OF INFORMATION SECTION === Device Model: HDS725050KLA360 Serial Number:KRVN02ZAG1SV7C Firmware Version: K2AOA11A User Capacity:500,107,862,016 bytes Device is:Not in smartctl database [for details use: -P showall] ATA Version is: 7 ATA Standard is: ATA/ATAPI-7 T13 1532D revision 1 Local Time is:Fri Oct 28 17:42:16 2005 EDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x82) Offline data collection activity was completed without error. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: (10419) seconds. Offline data collection capabilities:(0x5b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. No Conveyance Self-test supported. Selective Self-test supported. SMART capabilities:(0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability:(0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time:( 1) minutes. Extended self-test routine recommended polling time:( 174) minutes. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000b 100 100 016Pre-fail Always - 1 2 Throughput_Performance 0x0004 159 159 050Old_age Offline - 205 3 Spin_Up_Time0x0007 110 110 024Pre-fail Always - 618 (Average 645) 4 Start_Stop_Count0x0012 100 100 000Old_age Always - 34 5 Reallocated_Sector_Ct 0x0033 100 100 005Pre-fail Always - 0 7 Seek_Error_Rate 0x000a 100 100 067Old_age Always - 0 8 Seek_Time_Performance 0x0004 136 136 020Old_age Offline - 31 9 Power_On_Hours 0x0012 100 100 000Old_age Always - 691 10 Spin_Retry_Count0x0012 100 100 060Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000Old_age Always - 32 192 Power-Off_Retract_Count 0x0032 100 100 050Old_age Always - 62 193 Load_Cycle_Count0x0012 100 100 050Old_age Always - 62 194 Temperature_Celsius 0x0002 105 105 000Old_age Always - 52 (Lifetime Min/Max 22/62) 196 Reallocated_Event_Count 0x0032 100 100 000Old_age Always - 0 197 Current_Pending_Sector 0x0022 100 100
Re: Very slow writing to SATA disk
Ask the maintainer to get it [ataidle -mi] fixed, but be warned experience says it might hose your data... The maintainer did not break it. An incompatible change to the API did :) You are, probably, in the best position to show us, how the new API should be used. Now, you say read speed is OK, but write speed isnt, is that on the raw disk device or though the filesystem ? Everething is through the filesystem -- as stated in my original e-mail in this thread. There is no other activity, when a single cp reads the huge file from a SCSI disk (da1) to the IDE (ad8). According to `systat -vm', da1 is barely breaking a sweat, while ad8 is at 99-101% throughput at 7Mb/second. Soft-updates are on. The filesystem is almost empty. The box has a single dual-core Opteron-275 with 2Gb of RAM. The SATA controller is your favorite: atapci1: SiI 3114 SATA150 controller port 0xac00-0xac07,0xa480-0xa483,0xa400-0xa407,0xa080-0xa083,0xa000-0xa00f mem 0 xbe6fbc00-0xbe6fbfff irq 25 at device 5.0 on pci3 -mi ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
Look in smartmontools I provided patches for that, its not rocket science you know... This attitude -- on top of the API change itself -- is not really encouraging for ISVs, you know :-) You need to find out what the transfer rates are for the RAW disk, ie by doing a dd from /dev/zero to the disk with a a resonable blocksize say 1M to minimize overhead. Also read speed from disk to /dev/null blocksize 1M would be helpfull. The read test averaged 62603828 bytes/sec over a minute or so. The write test is only 6931231 bytes/sec -- about 9 times less. While dd is running, `systat 1 -vm' reports about 110 ata-interrupts (irq 25) per second during the write test, and about 1000 during the read test. Thanks for your help. -mi ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
Hmm, that does sound as problems with that disk, or maybe disk vs diskcontroller. Any chance you could try the disk on something else ? I'll try... One other thing, how much mem do you have in there ? more than 4G and bounce buffering might get into the picture ruining the transfer rate... 2Gb right now. Does the 4Gb limit apply to amd64 as well? -mi ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On Fri, Oct 28, 2005 at 07:09:55PM -0400, Mikhail Teterin wrote.. Hmm, that does sound as problems with that disk, or maybe disk vs diskcontroller. Any chance you could try the disk on something else ? I'll try... One other thing, how much mem do you have in there ? more than 4G and bounce buffering might get into the picture ruining the transfer rate... 2Gb right now. Does the 4Gb limit apply to amd64 as well? If your PCI card can only drive 32 addressbits you are toast beyond 4GB, in the sense that you need bounce buffers, regardless of the CPU. -- Wilko Bulte [EMAIL PROTECTED] ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On Fri, Oct 28, 2005 at 01:18:03AM -0400 I heard the voice of Mikhail T., and lo! it spake thus: According to smartctl, the drive runs at 56C during the copying. Its idle temperature seems to be 54C. That sounds a little high to me. Smartctl has been weird lately, and it only shows temp on one of my drives, but it (7200RPM 9gig SCSI) is showing 32C. The higher density probably burns a little more power, but half again the temperature? -- Matthew Fuller (MF4839) | [EMAIL PROTECTED] Systems/Network Administrator | http://www.over-yonder.net/~fullermd/ On the Internet, nobody can hear you scream. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On Fri, 2005-Oct-28 01:18:03 -0400, Mikhail T. wrote: I've got a new Hitachi drive with 16Mb of cache: ad8: 476940MB HDS725050KLA360 K2AOA11A at ata4-master SATA150 and am trying to use it to store backups online. Unfortunately, writing to the disk is painfully slow (by today's standards) -- it can barely keep 7Mb/second and my other (SCSI) disks run circles around it. If this drive doesn't support tagged-queueing, is the write cache disabled? I get that sort of performance from a (PATA) disk with the cache disabled (hw.ata.wc=0 in loader.conf) According to smartctl, the drive runs at 56C during the copying. Its idle temperature seems to be 54C. I'd double-check that (eg with a finger). If the drive really is running at 56°C, it won't last very long. -- Peter Jeremy ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
At 07:48 PM 10/28/2005 +1000, Peter Jeremy wrote: | On Fri, 2005-Oct-28 01:18:03 -0400, Mikhail T. wrote: | I've got a new Hitachi drive with 16Mb of cache: | | ad8: 476940MB HDS725050KLA360 K2AOA11A at ata4-master SATA150 | | and am trying to use it to store backups online. Unfortunately, writing | to the disk is painfully slow (by today's standards) -- it can barely | keep 7Mb/second and my other (SCSI) disks run circles around it. | | If this drive doesn't support tagged-queueing, is the write cache | disabled? I get that sort of performance from a (PATA) disk with | the cache disabled (hw.ata.wc=0 in loader.conf) | | According to smartctl, the drive runs at 56C during the copying. Its | idle temperature seems to be 54C. | | I'd double-check that (eg with a finger). If the drive really is | running at 56°C, it won't last very long. | | -- | Peter Jeremy I just checked the temp on my workstation for comparison. It's running two SATA drives (74GB Raptors) and the temp on those drives is 30C/86F. Like Peter says, at 133F, I don't think they would last very long, not to mention you'd probably smell something hot/burning. Ray ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On Fri, 2005-Oct-28 12:45:27 -0400, Mikhail Teterin wrote: No, just checked -- the hw.ata.wc is set to 1. Is there anything else to look at? That exhausts my ideas, sorry. Sören might be able to suggest something. According to smartctl, the drive runs at 56C during the copying. Its idle temperature seems to be 54C. It sure feels hot to the touch, but nothing is burning, A rule-of-thumb is that you can hold your finger on something for 4 seconds then it is 45°C (or less). this model's 'Ambient Temperature' spec is 5 to 55C... It's unlikely to affect I/O performance (though it may increase the number of thermal re-calibrations) but it will definitely shorten the drive life. -- Peter Jeremy ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On 28/10/2005, at 22:48, Peter Jeremy wrote: On Fri, 2005-Oct-28 12:45:27 -0400, Mikhail Teterin wrote: No, just checked -- the hw.ata.wc is set to 1. Is there anything else to look at? That exhausts my ideas, sorry. Sören might be able to suggest something. Not really, however I have no experience with Hitachi drives, I havn't trusted that line of drives since their name was IBM Deathstar DTLA series :/ According to smartctl, the drive runs at 56C during the copying. Its idle temperature seems to be 54C. It sure feels hot to the touch, but nothing is burning, A rule-of-thumb is that you can hold your finger on something for 4 seconds then it is 45°C (or less). this model's 'Ambient Temperature' spec is 5 to 55C... It's unlikely to affect I/O performance (though it may increase the number of thermal re-calibrations) but it will definitely shorten the drive life. Indeed, 55C is way to high for 24/7 usage, and it might be that the drive is choking on it and barely is able to compensate.. What does SMART say ? any unusual like high correction rates or anything ? Søren Schmidt [EMAIL PROTECTED] ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On 28/10/2005, at 23:45, Mikhail Teterin wrote: Indeed, 55C is way to high for 24/7 usage, and it might be that the drive is choking on it and barely is able to compensate. The reads are pretty quick... I'd like to be able to spin it down, but ataidle is broken :-( Ask the maintainer to get it fixed, but be warned experience says it might hose your data... What does SMART say ? any unusual like high correction rates or anything ? (SMART data deleted) Well except the excessive temperature nothing out of the ordinary... Now, you say read speed is OK, but write speed isnt, is that on the raw disk device or though the filesystem ? Søren Schmidt [EMAIL PROTECTED] ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On 29/10/2005, at 0:03, Mikhail Teterin wrote: Ask the maintainer to get it [ataidle -mi] fixed, but be warned experience says it might hose your data... The maintainer did not break it. An incompatible change to the API did :) You are, probably, in the best position to show us, how the new API should be used. Look in smartmontools I provided patches for that, its not rocket science you know... Now, you say read speed is OK, but write speed isnt, is that on the raw disk device or though the filesystem ? Everething is through the filesystem -- as stated in my original e- mail in this thread. There is no other activity, when a single cp reads the huge file from a SCSI disk (da1) to the IDE (ad8). According to `systat -vm', da1 is barely breaking a sweat, while ad8 is at 99-101% throughput at 7Mb/second. Soft-updates are on. The filesystem is almost empty. The box has a single dual-core Opteron-275 with 2Gb of RAM. The SATA controller is your favorite: atapci1: SiI 3114 SATA150 controller port 0xac00-0xac07,0xa480-0xa483,0xa400-0xa407,0xa080-0xa083,0xa000-0xa00f mem 0 xbe6fbc00-0xbe6fbfff irq 25 at device 5.0 on pci3 OK, then we dont know where the slowdown is yet... You need to find out what the transfer rates are for the RAW disk, ie by doing a dd from /dev/zero to the disk with a a resonable blocksize say 1M to minimize overhead. Also read speed from disk to /dev/null blocksize 1M would be helpfull. Søren Schmidt [EMAIL PROTECTED] ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On 29/10/2005, at 0:41, Mikhail Teterin wrote: Look in smartmontools I provided patches for that, its not rocket science you know... This attitude -- on top of the API change itself -- is not really encouraging for ISVs, you know :-) Sigh, ataidle is a hack and the author had no intention to listen back when, so I dont feel teribly sorry about it you know. Spinning down disks needs to be done at the driver level so ATA knows what state the disk is in etc... You need to find out what the transfer rates are for the RAW disk, ie by doing a dd from /dev/zero to the disk with a a resonable blocksize say 1M to minimize overhead. Also read speed from disk to /dev/null blocksize 1M would be helpfull. The read test averaged 62603828 bytes/sec over a minute or so. The write test is only 6931231 bytes/sec -- about 9 times less. While dd is running, `systat 1 -vm' reports about 110 ata- interrupts (irq 25) per second during the write test, and about 1000 during the read test. Hmm, that does sound as problems with that disk, or maybe disk vs diskcontroller. Any chance you could try the disk on something else ? One other thing, how much mem do you have in there ? more than 4G and bounce buffering might get into the picture ruining the transfer rate... Søren Schmidt [EMAIL PROTECTED] ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
On 29/10/2005, at 1:09, Mikhail Teterin wrote: Hmm, that does sound as problems with that disk, or maybe disk vs diskcontroller. Any chance you could try the disk on something else ? I'll try... One other thing, how much mem do you have in there ? more than 4G and bounce buffering might get into the picture ruining the transfer rate... 2Gb right now. Does the 4Gb limit apply to amd64 as well? Yes, the SiI3114 is a 32bit device.. Søren Schmidt [EMAIL PROTECTED] ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Very slow writing to SATA disk
Søren Schmidt wrote: On 28/10/2005, at 23:45, Mikhail Teterin wrote: Indeed, 55C is way to high for 24/7 usage, and it might be that the drive is choking on it and barely is able to compensate. The reads are pretty quick... I'd like to be able to spin it down, but ataidle is broken :-( Ask the maintainer to get it fixed, but be warned experience says it might hose your data... What does SMART say ? any unusual like high correction rates or anything ? (SMART data deleted) Well except the excessive temperature nothing out of the ordinary... Now, you say read speed is OK, but write speed isnt, is that on the raw disk device or though the filesystem ? Søren Schmidt [EMAIL PROTECTED] For what it's worth, I'm seeing slow write speeds on some tests with other (non-ata) controllers. Haven't had time to isolate it just yet. Scott ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]