Page 2 of 4 FirstFirst 1234 LastLast
Results 11 to 20 of 35

Thread: System->Admin->Disk Utility->Benchmark results

  1. #11
    Join Date
    Dec 2006
    Location
    Melbourne, Australia
    Beans
    431

    Re: System->Admin->Disk Utility->Benchmark results

    Quote Originally Posted by Tweak42 View Post
    Thanks for pointing out the nifty bench utility feature.

    Your drive may be showing up under IDE mode if you don't have the controller set to ahci/sata in the bios. Just beware switching modes can halt a WinXP partition from booting; linux, vista and win7 seem handle switching it ok.
    Thank you for this info.

    I has wondered why I was getting error messages "ATA1 device not ready. Softreset failed" on bootup.

    Changing over to ahic/sata fixed that and although Ubuntu could read the Win XP partitions XP did not boot. I see a re-install of Win XP coming up! Sigh!

  2. #12
    Join Date
    Nov 2004
    Location
    Osnabrück, Germany
    Beans
    143

    Re: System->Admin->Disk Utility->Benchmark results

    250GB WD Caviar Blue WD2500AAKS-00B3A0 (ext4):

    Read min: 41.8 MB/s
    Read max: 97.1 MB/s
    Read avg: 75.4 MB/s
    Access avg: 16.5 ms

    1TB WD Caviar Green WD10EADS-00L5B1 (ext3):

    Read min: 44.3 MB/s
    Read max: 92.0 MB/s
    Read avg: 73.9 MB/s
    Access avg: 14.1 ms

    BTW: It would be cool if we could have some sort of Database like HD Tune has.

    Edit: Both hdds are completely encrypted if that means anything.
    Attached Images Attached Images
    Last edited by buellman; April 28th, 2010 at 12:49 PM.

  3. #13
    Join Date
    Nov 2004
    Location
    Osnabrück, Germany
    Beans
    143

    Re: System->Admin->Disk Utility->Benchmark results

    Quote Originally Posted by andrewabc View Post
    Can anyone confirm if safe to do read/write benchmark? Don't want to screw any of my stuff up.
    Try it It will tell you the following:
    A partition table was detected - write benchmarking requires the disk to be completely empty

  4. #14
    Join Date
    Oct 2007
    Beans
    Hidden!

    Re: System->Admin->Disk Utility->Benchmark results

    As someone else already asked: is the read/write test destructive? Would be nice if they made that more clear!
    (edit already answered above)

    Anyway, rather than bragging about how fast my raided ssds are, lets see who can post a slower result than this:

    Min read 10.3 MB/s (!)
    Max read 30.7 MB/s
    Avg read 24.8 MB/s

    avg access time: 17.8ms

    Thats using a prehistoric 40GB notebook drive. (and interestingly, ubuntu is still rather snappy on it).


  5. #15
    Join Date
    Dec 2005
    Beans
    1,171
    Distro
    Ubuntu 10.10 Maverick Meerkat

    Re: System->Admin->Disk Utility->Benchmark results

    Quote Originally Posted by buellman View Post
    Try it It will tell you the following:
    Yes, I noticed that for HDD.
    For my SSD it only asked if I was sure I wanted to do the read/write. No mention of it needing to be empty or a partition.
    I didn't do it because I figured it would corrupt my SSD install. A better warning would be appreciated as someone is likely to run the test because not enough warning given.
    Last edited by andrewabc; April 28th, 2010 at 02:16 PM.

  6. #16
    Join Date
    Oct 2007
    Beans
    Hidden!

    Re: System->Admin->Disk Utility->Benchmark results

    Quote Originally Posted by andrewabc View Post
    Yes, I noticed that for HDD.
    For my SSD it only asked if I was sure I wanted to do the read/write. No mention of it needing to be empty or a partition.
    I didn't do it because I figured it would corrupt my SSD install. A better warning would be appreciated as someone is likely to run the test because not enough warning given.
    Hmm.. an ssd doesnt have an inner and outer sector and so you wouldnt need to write (or read) across the entire drive to bench it. So quite possibly it just needs some free space on a ssd to test at random?

  7. #17
    Join Date
    Aug 2006
    Location
    Smederevo/Serbia
    Beans
    175
    Distro
    Ubuntu 14.04 Trusty Tahr

    Re: System->Admin->Disk Utility->Benchmark results

    Western Digital Caviar Green 1TB WD10EADS-00L5B1
    Western Digital Caviar Blue 640GB WD6400AAKS-22A7B0
    Western Digital Caviar Blue 320GB WD3200AAKS-00VYA0
    Maxtor 160GB 6Y160P0

    All drives use ext4 file system.
    Last edited by Milos_SD; April 28th, 2010 at 05:10 PM.

  8. #18
    Join Date
    Apr 2007
    Beans
    307
    Distro
    Ubuntu 12.04 Precise Pangolin

    Re: System->Admin->Disk Utility->Benchmark results

    The graphical display of the results of the benchmark, I think are not displayed correctly.

    What may be more meaningful is having collected the data (read rates and access times) is to sort them in increasing order. That way the median (i.e at 50% of the time) would be the average. The lower and upper quartiles (and/or deciles) could also be determined from the dataset.

    The upper quartile would then represent the read rate at which 25% of the data exceeds that amount.

    The minimum & maximum values are less informative because they may have occurred for an insignificant amount of time. For example, if you have a maximum read rate of 200MB/s and it is the only reading above 100MB/s, then it isn't particularly meaningful.

    The access times should be displayed in a similar way.
    'I am writing to you, knowing that you will do even more than I ask.'

  9. #19
    Join Date
    Dec 2005
    Beans
    1,171
    Distro
    Ubuntu 10.10 Maverick Meerkat

    Re: System->Admin->Disk Utility->Benchmark results

    Quote Originally Posted by Onesimus View Post
    The graphical display of the results of the benchmark, I think are not displayed correctly.

    What may be more meaningful is having collected the data (read rates and access times) is to sort them in increasing order. That way the median (i.e at 50% of the time) would be the average. The lower and upper quartiles (and/or deciles) could also be determined from the dataset.

    The upper quartile would then represent the read rate at which 25% of the data exceeds that amount.

    The minimum & maximum values are less informative because they may have occurred for an insignificant amount of time. For example, if you have a maximum read rate of 200MB/s and it is the only reading above 100MB/s, then it isn't particularly meaningful.

    The access times should be displayed in a similar way.
    I didn't notice any major problems. The average would be average for the whole test, not just an average between min/max. Although a 200mb datapoint even though nothing else is above 100mb/s does throw off avg a bit. The way to solve this would be to use statistical analysis to eliminate irregular occurrences (forget proper term) from calculating the averages. But I wouldn't prevent min/max from being shown.

    Min and max are good info, true the min/max # might only occur for extremely short period of time and rarely occur, but does show the min/max possible during the test.

    Don't see what is wrong with avg latency. Just gives an average of all points. HDD that number is all over the place. SSD it stays pretty much in same spot the whole test.

    I agree the program can be improved a lot. It definitely has been improved a lot since 9.10. Hopefully more progress can be made for 10.10

  10. #20
    Join Date
    Mar 2006
    Location
    Williams Lake
    Beans
    Hidden!
    Distro
    Ubuntu Development Release

    Re: System->Admin->Disk Utility->Benchmark results

    I've got a Mac G3 running Debian stable PPC, It doesn't have a gui, so here are the hdparm results:

    Code:
    /dev/hda:
     Timing cached reads:   358 MB in  2.00 seconds = 178.61 MB/sec
     Timing buffered disk reads:   60 MB in  3.08 seconds =  19.49 MB/sec
    Model=WDC WD102AA, FwRev=83.10A83

Page 2 of 4 FirstFirst 1234 LastLast

Bookmarks

Posting Permissions

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