hitachi 7k1000.c 4x500g raid 0 performance

S

SinShiva

Limp Gawd
Joined
Nov 20, 2009
Messages
330
thought this might make a good reference.

sleepy_raid.png


ICH10R chipset on the msi x58m mobo, i7 [email protected] with 12gb of ddr3 1600

was hoping somebody with a similar setup using samsung f3's could post to compare?

the raid volume is 200gb with 64kb stripe size. write cache enabled

http://www.newegg.com/Product/Product.aspx?Item=N82E16822145299

those are the hdds i bought. the pricing is comparable to the 500gb samsung f3s, but the availability is much nicer :p

i took a chance, but knowing the 7k1000.c used 500gb platters was enough for me. i am rather pleased with this investment.
 
If possible could you kindly perform some tests to help/confirm for all the TLER-dependent users is it possible to set TLER, or in Hitachi's term CCTL on those 7K1000.C drives? I just purchased 8 WD drives (partial no TLER, OK for me now) earlier this month and considering purchasing more in the next 2 months. if Hitachi is good I think a lot will consider if cost/performance is reasonable.

Thank you for your kind help.
 
i tried to run hdat2, but it couldn't seem to find drivers for my system. might be that i was running the disk from an external dvdrom, etc. i spent some time looking for CCTL utilities for hitachi and have come up with zip. any suggestions? i'm more than willing to try, i just can't seem to find something will work with hitachi. now that you've mentioned it and i've read up on it, i'd very much like to enable it as well.
 
If HDAT2 is not working for you, try the following way. It may not confirm but I need the initial info.

1. If it is ok for you, find an updated Linux LiveCD that you think has proper driver for your storage controller card. This is a bit difficult because the version must have package "smartmontools" integrated. Some newer LiveCD (i think Fedora 12 live) removed that package.

2. Boot from Linux LiveCD with smartmontools, if storage controller detected properly, run the shell command smartctl --all /dev/sda, reports everything related to SCT capabilities

Thank you for your kind attention and assistance.
 
[root@localhost liveuser]# smartctl --all /dev/sda


smartctl version 5.38 [i386-redhat-linux-gnu] Copyright (C) 2002-8 Bruce Allen

Home page is http://smartmontools.sourceforge.net/



=== START OF INFORMATION SECTION ===

Device Model: Hitachi HDS721050CLA362

Serial Number: JP1511HN035AMA

Firmware Version: JP2OA25C

User Capacity: 500,107,862,016 bytes

Device is: Not in smartctl database [for details use: -P showall]

ATA Version is: 8
ATA Standard is: ATA-8-ACS revision 4
Local Time is: Wed Dec 30 01:41:13 2009 EST

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: (0x84)
Offline data collection activity was suspended by an interrupting command from host.

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: (4890) 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: ( 82) minutes.

SCT capabilities: (0x003d) SCT Status supported.

SCT Feature Control supported.

SCT Data Table supported.


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 100 100 016 Pre-fail Always - 0
2
Throughput_Performance 0x0005 135 135 054 Pre-fail Offline - 97
3
Spin_Up_Time 0x0007 121 121 024 Pre-fail Always - 188 (Average 191)
4
Start_Stop_Count 0x0012 100 100 000 Old_age Always - 161

Reallocated_Sector_Ct 0x0033 100 100 005 Pre-fail Always - 0
7
Seek_Error_Rate 0x000b 100 100 067 Pre-fail Always - 0
8
Seek_Time_Performance 0x0005 132 132 020 Pre-fail Offline - 34
9
Power_On_Hours 0x0012 100 100 000 Old_age Always - 43
10
Spin_Retry_Count 0x0013 100 100 060 Pre-fail Always - 0
12
Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 160
192
Power-Off_Retract_Count 0x0032 100 100 000 Old_age Always - 161
193
Load_Cycle_Count 0x0012 100 100 000 Old_age Always - 161
194
Temperature_Celsius 0x0002 153 153 000 Old_age Always - 39 (Lifetime Min/Max 25/45)
196
Reallocated_Event_Count 0x0032 100 100 000 Old_age Always - 0
197
Current_Pending_Sector 0x0022 100 100 000 Old_age Always - 0
198
Offline_Uncorrectable 0x0008 100 100 000 Old_age Offline - 0
199
UDMA_CRC_Error_Count 0x000a 200 200 000 Old_age Always - 0


SMART Error Log Version: 0

No Errors Logged



SMART Self-test log structure revision number 1

No self-tests have been logged. [To run self-tests, use: smartctl -t]



SMART Selective self-test log data structure revision number 1

SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS

1 0 0 Not_testing

2 0 0 Not_testing

3 0 0 Not_testing

4 0 0 Not_testing

5 0 0 Not_testing


Selective self-test flags (0x0):

After scanning selected spans, do NOT read-scan remainder of disk.

If Selective self-test is pending on power-up, resume after 0 minute delay.
Click to expand...

[text formatting fucked up, sorry.]

Cheers. :) - is there a way to pass the SMART commands from linux and have it stick permanently ? regarding CCTL, i mean
 
Last edited:
Hi SinShiva

1. TLER/CCTL relevant description under ATAPI command set reference document is called Error Recovery Control Command. Usually grouped under SMART SCT capabilities.

2. The posted report is a summary of generally visible information available to normal disk software. Generic hard disk can be accessed thorough standard known method such as SMART-aware disk/tool interface. They can also be directly managed by disk vendor low-level tools manipulating the bits of the firmware images.

3. I wanted to know the SCT because HDAT2 is detecting for ERCC under the SCT menu. So this is a start.

4. My apology at this point of time my knowledge is not enough to extract/decipher more info through the Linux environment. However, I am now confident you should be able to use HDAT2 to gain more info.

5. [Only do this if you are comfortable]
5a. boot into BIOS
5b. switch disk to ATA/Native-IDE mode temporary. save and reboot
5c. boot into dos-floppy or equivalent containing the hdat2 exceutable.
5d. example run : hdat2 /d=1,2,10 /s /x=1 /t=2,1
5e. if everything nice, HDAT2 now presents you the detected Hard Disks. Select and go in further to explore the info related to that hard disk. You want to look for SMART. Be very careful with the tool because HDAT2 has direct interface access to many low-level feature/functions and even ATA command sets.
5f. If available, you will eventually find a SCT submenu something like Error Recovery Control, you will have 2 settings, one is for READ and another one for WRITE. Do not change anything, just let us know the default value for now. I want to avoid you changing the wrong value leading to problem on your current array.
5g. quit and reboot back to BIOS, change your disk mode to original working setting.

I hope you can give us good news ! Cheers.
 
SCT version: 0100h
SCT level: 0001h


the value is 0 for both read and write, should i set this to 7 for all 4 hdds?

they both can be changed, but i'd rather wait for a go ahead from you
 
1. This is a very good start. I appreciate your effort. This implies that at the very least, for this particular Model with the firmware revision you have on the disk itself, the Error Recovery Control Command feature is available as advertised in Hitachi datasheet.

The following is important to others users
WD prints the model and firmware revision and manufacturing date on the disk label so you can actually physically verify.
I am not sure about Hitachi. Hitachi may use the same drive but load different firmware and prints only the model code on the disk label. Hitachi users can tell us whether you could identify just by looking at the disk label and find respective firmware revision.

So in summary for point 1, you know the feature is present in your own setup.

2. What to do next? This is slightly more complicated because
2a. Some users said while you could modify the values using HDAT2, the value cannot be retained upon power cycle.
2a-1 In the old case of WDTLER, the program knows how to set a permanent value on supported WD drives.
2a-2 User unable to find official Hitachi tools to set this on permanent basis.

I will suggest goto more knowledgable users to lead you on Hitachi-side idea because I do not have enough info on CCTL tools.

I will experiment further with HDAT2 or other tools. However since disk vendors do not support enduser playing with low-level disk parameters or firmware-level modification, my current knowledge is not sufficient so I feel it is not appropriate for me to comment too much on this for fear of potential damage and data loss that could occur if others follows blindly without due preparation/protection. If you must ask, perhaps directly to the authors of disk tools.

Just know that in future if a supported permanent solution is found, your setup is capable of making the necessary change.

Cheers.
 
For what is worth, SinShiva, the following is really meant for others who can make sense out of it. I do not belong to that group. Minimum CCTL is 50ms. Usually you need more than that.

(Default CCTL) =
((content of the Features register)*(Identify Device words (99:98))) micriseconds.

(Yes :) "micriseconds" the exact word I found in the specification)
 
OK SinShiva. This is the final effort and update I can contribute before calling it a day.

This post is very specific to latest consumer SATA hard disks that consumer wanted to try to use for SOHO RAID.

1. If the drive firmware already removed the TLER/CCTL capability. This post does not have any immediate help.

2. This post requires complicated steps and I guaranteed this post does not give you final answer, only road further ahead.

3. This post is specific to hard disk that has CCTL/TLER or similar capability but you do not have easy way to access them. (you still do not have easy way after this post, I admit)

4. Most users cannot benefit from this post, even myself. But I felt I owe SinShiva at least an effort and a proper reply after asking SinShiva to do some research works and raising hope along the way.

OK here it goes,

1. Due to software reason, you cannot even start with this specific suggestion if you are using motherboard built-in SATA controllers. This include all standard Intel/AMD/NVIDIA types.

2. It is not because the hardware cannot support such a request,(I am not software person so I feel need to give reasonable buffer zone for the following) it is mostly due to the requested function not made available/merged into commonly available tools that will make it easy for consumer-usage. This perhaps is not bad because you definitely do not want consumer to play with low-level disk settings.

3. There is a possible alternative for the very specific scenario I repeatedly clarify above.

4. You need a SAS card to try. The specific disk connected to the SAS card. This is the reason why I was stuck because personally I do not have any SAS card at home.

5. You must run from the Linux environment. This also means Windows only user unfortunately I just don't have any way. The tool has Windows port but I recommend extreme caution because it was originally written for Unix environment and you really do not want to send the wrong command to disk. Perhaps it will be easier from BSD but I do not have energy nor resource to recreate a separate BSD machine, virtual machine will not help in this case, and I do not have a live BSD CD as well. Even if you do, the needed tool is usually not part of standard packages.

6. If everything aligned, you can try Linux sdparm to issue the relevant command. You may need additional software packages. If you are successful, this will solve the problem of reset upon power-cycle, either updating final value or re-issue the relevant command during start-up. This command and related utilities are very complex and dealing with low-level disk settings and operations, issuing the wrong command can be very harmful. For the record I do NOT recommend consumer to even try this. I will not be responsible for any problems and I will not answer any questions related to it.

Remember:
I did not complete test because I do not have SAS card at home. This means I have NO confirmation about further progress and I am not sure about correctness of this post. Verify with all knowledgeable professionals before considering anything. You have been warned.

Cheers
 
You must log in or register to reply here.
Back
Top