Results 1 to 21 of 21
  1. #1
    Join Date
    Aug 2011
    Posts
    141

    Usual results for RAID 10

    I was wondering if these were correct results, now I know it's not a valid I/O test but it gives a rough idea. It is 4 SATA drives running in software RAID 10.

    Code:
    hdparm -tT /dev/sda
    
    /dev/sda:
     Timing cached reads:   6260 MB in  2.00 seconds = 3131.80 MB/sec
     Timing buffered disk reads:  216 MB in  3.01 seconds =  71.64 MB/sec


    Code:
    dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 9.81648 seconds, 109 MB/s

    I have seen much much better results from other servers running RAID 10 with SATA drives.

  2. #2
    Join Date
    Aug 2011
    Posts
    141
    Anyone has any input on the matter?

  3. #3
    Join Date
    Feb 2010
    Location
    East Coast
    Posts
    410
    What hard drives are you using?

    You should be able to get 140 MB/s with newer HDDs.

  4. #4
    Join Date
    Aug 2011
    Posts
    141
    Quote Originally Posted by BassHost View Post
    What hard drives are you using?

    You should be able to get 140 MB/s with newer HDDs.
    I got told they were brand new HITACHI drives.

  5. #5
    Join Date
    Jan 2007
    Location
    Germany
    Posts
    198
    Is there running some sort of rebuild of your software raid or other i/o load?

    Which model are the drives?
    Just that they are new doesn't say anything about their performance:
    - 5400 rpm drives
    - 2.5" vs. 3.5" drives
    - new drive, but still an old model

  6. #6
    Join Date
    Apr 2003
    Location
    Los Angeles, CA
    Posts
    820
    That doesn't look very good. For comparison, here's a 3 year old server with 4 x 1 TB WD SATA drives (also 3 years old) in RAID10:
    Code:
    Timing cached reads:   11028 MB in  2.00 seconds = 5520.62 MB/sec
    Timing buffered disk reads:  830 MB in  3.00 seconds = 276.43 MB/sec
    Pings <1 ms, Unlimited Transfer, Lowest Price: http://localhost/

  7. #7
    Join Date
    Jan 2007
    Location
    /dev/null
    Posts
    3,700
    Here is a VM result of my personal test environment (RAID 10) with 4 normal SATA drives:
    Code:
    1073741824 bytes (1.1 GB) copied, 6.86996 s, 156 MB/s
    Code:
     Timing cached reads:   14094 MB in  2.00 seconds = 7055.56 MB/sec
     Timing buffered disk reads:  1076 MB in  3.00 seconds = 358.47 MB/sec
    From:
    http://www.webhostingtalk.com/showth...55#post8005955

    Yours is indeed really low as luki pointed out.

  8. #8
    Join Date
    Aug 2011
    Posts
    141
    I did think it was low and I was very disappointed when I was told it was the software RAID causing that issue.


    Drive information:

    Code:
    hdparm -I /dev/sda
    
    /dev/sda:
    
    ATA device, with non-removable media
            Model Number:       ST3500630NS
            Serial Number:      9QG84ESY
            Firmware Revision:  3BKS
    Standards:
            Supported: 7 6 5 4
            Likely used: 7
    Configuration:
            Logical         max     current
            cylinders       16383   16383
            heads           16      16
            sectors/track   63      63
            --
            CHS current addressable sectors:   16514064
            LBA    user addressable sectors:  268435455
            LBA48  user addressable sectors:  976773168
            device size with M = 1024*1024:      476940 MBytes
            device size with M = 1000*1000:      500107 MBytes (500 GB)
    Capabilities:
            LBA, IORDY(can be disabled)
            Queue depth: 32
            Standby timer values: spec'd by Standard, no device specific minimum
            R/W multiple sector transfer: Max = 16  Current = ?
            Advanced power management level: unknown setting (0xfefe)
            Recommended acoustic management value: 208, current value: 0
            DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6
                 Cycle time: min=120ns recommended=120ns
            PIO: pio0 pio1 pio2 pio3 pio4
                 Cycle time: no flow control=240ns  IORDY flow control=120ns
    Commands/features:
            Enabled Supported:
               *    SMART feature set
                    Security Mode feature set
               *    Power Management feature set
               *    Write cache
               *    Look-ahead
               *    Host Protected Area feature set
               *    WRITE_BUFFER command
               *    READ_BUFFER command
               *    DOWNLOAD_MICROCODE
               *    Advanced Power Management feature set
                    SET_MAX security extension
                    Automatic Acoustic Management feature set
               *    48-bit Address feature set
               *    Device Configuration Overlay feature set
               *    Mandatory FLUSH_CACHE
               *    FLUSH_CACHE_EXT
               *    SMART error logging
               *    SMART self-test
               *    General Purpose Logging feature set
               *    64-bit World wide name
                    unknown 84[11]
                    unknown 84[12]
               *    SATA-I signaling speed (1.5Gb/s)
               *    SATA-II signaling speed (3.0Gb/s)
               *    Native Command Queueing (NCQ)
               *    Phy event counters
                    Device-initiated interface power management
               *    Software settings preservation
    Security:
            Master password revision code = 65534
                    supported
            not     enabled
            not     locked
            not     frozen
            not     expired: security count
            not     supported: enhanced erase
    Checksum: correct

    Code:
    smartctl -i /dev/sda
    smartctl version 5.38 [x86_64-redhat-linux-gnu] Copyright (C) 2002-8 Bruce Allen
    Home page is http://smartmontools.sourceforge.net/
    
    === START OF INFORMATION SECTION ===
    Model Family:     Seagate Barracuda ES
    Device Model:     ST3500630NS
    Serial Number:    9QG84ESY
    Firmware Version: 3BKS
    User Capacity:    500,107,862,016 bytes
    Device is:        In smartctl database [for details use: -P show]
    ATA Version is:   7
    ATA Standard is:  Exact ATA specification draft version not indicated
    Local Time is:    Mon Mar 12 01:41:52 2012 CDT
    SMART support is: Available - device has SMART capability.
    SMART support is: Enabled

  9. #9
    Join Date
    Jan 2007
    Location
    /dev/null
    Posts
    3,700
    2006, is this a production server, or are you able to play with it? As I'm curious what the performance per disk is (maybe 1 is really slow, dragging the rest down with it).

  10. #10
    Join Date
    Aug 2011
    Posts
    141
    Quote Originally Posted by Robert vd Boorn View Post
    2006, is this a production server, or are you able to play with it? As I'm curious what the performance per disk is (maybe 1 is really slow, dragging the rest down with it).
    I actually requested a refund on it and was waiting for billing to reply to be today so I have access to it and it's not a production server.

    I am not aware of the speed of the drives as the host did not state which is my fault as I should have checked.

  11. #11
    Join Date
    Jan 2007
    Location
    /dev/null
    Posts
    3,700
    I guess it doesn't come with a KVM/IPMI device? It would be nice to just do a quick CentOS reinstall (~10 minutes) and perform some single-disk tests.

  12. #12
    Join Date
    Aug 2011
    Posts
    141
    It does on request, I'll just say it's with datashack and this is the first main 'disappointment' I've had with them and I have 7 servers. I was looking into branching out to RAID 10 and replacing the existing servers I have with new RAID 10 configurations but from what I've seen I am not so sure now.

    I wouldn't be bothered to ask for KVM as I am expecting a reply from billing soon as it is now monday where I am so I assume they're just waking up.

  13. #13
    Maybe they gave you very old drives? I have a server with them and i got provisioned a drive with almost 30.000 hours with a few bad sectors, so far it works fine and i really don't care too much as im not storing important data, check the SMART report of the drives to see how healty are they.

  14. #14
    Join Date
    Aug 2011
    Posts
    141
    Quote Originally Posted by Lambda View Post
    Maybe they gave you very old drives? I have a server with them and i got provisioned a drive with almost 30.000 hours with a few bad sectors, so far it works fine and i really don't care too much as im not storing important data, check the SMART report of the drives to see how healty are they.
    Mentioned earlier that I was told they were brand new.

  15. #15
    Join Date
    Nov 2006
    Location
    Pune, India
    Posts
    1,428
    hdparm -tT /dev/sda

    You are checking only 1 drive using this. To check for RAID10 , check on /dev/mdX on the / partition.

    Also check whether software RAID has completed the sync

    cat /proc/mdstat

    For a single drive 71 MB/s is a bit low, you would generally expect ~ 100 MB/s with nothing running on it.

    Ishan
    LeapSwitch Networks Pvt. Ltd. - Managed VPS / Dedicated Servers India
    CloudJiffy PaaS - Wordpress Cluster Hosting
    █ Shared, Reseller, VPS, Dedicated Servers, Colocation
    AS132335 - India - USA - Germany - Spain - Portugal - Ukraine

  16. #16
    Join Date
    Jan 2008
    Location
    Europe
    Posts
    779
    Quote Originally Posted by mcdl View Post
    I got told they were brand new HITACHI drives.
    Quote Originally Posted by mcdl View Post
    Model Number: ST3500630NS
    ST3500630NS = Seagate first produced in 2006, most places stopped stocking them years ago (eg, newegg) and I doubt they're new, although it is possible.
    Last edited by sam0; 03-12-2012 at 08:51 AM.

  17. #17
    Join Date
    Aug 2011
    Posts
    141
    Code:
    hdparm -tT /dev/md0
    
    /dev/md0:
     Timing cached reads:   11772 MB in  2.00 seconds = 5892.47 MB/sec
     Timing buffered disk reads:  706 MB in  3.00 seconds = 235.25 MB/sec
    dd still low though, lower than the results I've seen.

  18. #18
    Join Date
    Nov 2006
    Location
    Pune, India
    Posts
    1,428
    Quote Originally Posted by mcdl View Post
    Code:
    hdparm -tT /dev/md0
    
    /dev/md0:
     Timing cached reads:   11772 MB in  2.00 seconds = 5892.47 MB/sec
     Timing buffered disk reads:  706 MB in  3.00 seconds = 235.25 MB/sec
    dd still low though, lower than the results I've seen.
    This looks much better.

    Do hdparm on each drive separately and also run -

    smartctl -t long /dev/sdX on each drive

    You can check result using -

    smartctl -l selftest /dev/sdX after the test completes.

    You can check whether test is complete, -

    smartctl -c /dev/sdX
    LeapSwitch Networks Pvt. Ltd. - Managed VPS / Dedicated Servers India
    CloudJiffy PaaS - Wordpress Cluster Hosting
    █ Shared, Reseller, VPS, Dedicated Servers, Colocation
    AS132335 - India - USA - Germany - Spain - Portugal - Ukraine

  19. #19
    Join Date
    Mar 2012
    Posts
    162
    With two 10K 3GBS Sata drives in Raid 0 I see about 175 MB/s with that same DD test.

  20. #20
    Join Date
    Jan 2007
    Location
    /dev/null
    Posts
    3,700
    Quote Originally Posted by ishan View Post
    hdparm -tT /dev/sda

    You are checking only 1 drive using this. To check for RAID10 , check on /dev/mdX on the / partition.
    That's due to it being software RAID then? Will help in the future, thanks (physical disk vs partition I guess?).

  21. #21
    Join Date
    Nov 2006
    Location
    Pune, India
    Posts
    1,428
    Quote Originally Posted by Robert vd Boorn View Post
    That's due to it being software RAID then? Will help in the future, thanks (physical disk vs partition I guess?).
    Yes, this is due to software RAID.

    In Hardware RAID you will probably see just one /dev/sda as OS will see it as a single drive.
    LeapSwitch Networks Pvt. Ltd. - Managed VPS / Dedicated Servers India
    CloudJiffy PaaS - Wordpress Cluster Hosting
    █ Shared, Reseller, VPS, Dedicated Servers, Colocation
    AS132335 - India - USA - Germany - Spain - Portugal - Ukraine

Similar Threads

  1. Script to Sort Databse Results from Results Page
    By Searching10 in forum Software & Scripts Requests
    Replies: 4
    Last Post: 10-01-2010, 02:03 AM
  2. 400GB Hard Disk Drives in RAID 0, RAID 5 and RAID 10 Arrays: Performance Analysis
    By donniesd in forum Hosting Security and Technology
    Replies: 0
    Last Post: 03-07-2007, 03:19 PM
  3. Is it usual...
    By csparks in forum Dedicated Server
    Replies: 13
    Last Post: 04-12-2006, 05:22 PM
  4. Usual Time
    By SearchNetNow in forum Domain Names
    Replies: 3
    Last Post: 12-31-2004, 11:13 PM
  5. 30% off as usual
    By cywkevin in forum Shared Hosting Offers
    Replies: 0
    Last Post: 06-30-2004, 12:11 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •