Announcement

Collapse
No announcement yet.

EXT3, EXT4, Btrfs Ubuntu Netbook Benchmarks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Hi I'm not Canonical

    Michael:

    I thought from previous articles I read Ext4 had regressed substantially.

    Also did you ever pin down what caused the data loss from the video card tests?

    Any word out of Google about their move to ext4?

    Comment


    • #32
      Originally posted by mtippett View Post
      The intent of the test is to show the FS performance, and to simulate some contrived situations where people can't do transactions (like a busy 2-tier PHP site. The 3rd tier allows you to aggregate transactions, but a 2-tier won't)..
      I'm not sure what you exactly mean by those tiers, but if you mean lots of people with lots of databases (or even database engines) on the same machine then your sqlite test is flawed - it performs many serialized micro transactions instead of many parallel macro transactions. (I hope you can see the difference and the conclusions yourself.)

      Comment


      • #33
        Originally posted by cl333r View Post
        I wouldn't use ntfs for storing data for it isn't posix compatible so you loose meta info (like whether the file is executable or not) when storing files, not to mention that writing to ntfs (for me) is (a lot) slower than to ext4 under Linux. Also, it's enemy's territory. Can't wait for the final btrfs and see whether it delivers what it promises.
        Plus cpu usage goes way up when writing/reading large amounts of data (or perhaps just large files) from an ntfs disk. At least it did in Jaunty. I've since moved all my ntfs-housed data to my ext4 partition.

        Comment


        • #34
          So might one recommend btrfs over ext4 for lucid?

          Comment


          • #35
            Originally posted by molecule-eye View Post
            So might one recommend btrfs over ext4 for lucid?
            What do you have against plain old xfs or jfs?

            Comment


            • #36
              Originally posted by phtpht View Post
              What do you have against plain old xfs or jfs?
              Why would you think I have anything against them? I was just wondering, given the results of the article, if anyone would recommend btrfs over ext4. What do you have against kaleidoscopes?

              Comment


              • #37
                So, in practical terms, is btrfs good enough to be used instead of ext4? Is it stable and does it keep data correctly? Also, just out of curiosity, do we know the reasons behind the various regressions and progressions?

                Comment


                • #38
                  Originally posted by atmartens View Post
                  So, in practical terms, is btrfs good enough to be used instead of ext4? Is it stable and does it keep data correctly? Also, just out of curiosity, do we know the reasons behind the various regressions and progressions?
                  If you care about your data, not trust any new FS. btrfs isn't marked "stable" in the kernel. ext4 is "stable" but some users have had data losses. I use ext4 for my root FS (I can always use a LiveCD/USB to have a OS), but for data I use ext3.

                  Comment


                  • #39
                    The SQLite test as it stands is *very* sensitive to the fsync/fdatasync. You can tune the test to demonstrate database performance, but it currently isn't intended to do that. The intent of the test is to show the FS performance, and to simulate some contrived situations where people can't do transactions (like a busy 2-tier PHP site. The 3rd tier allows you to aggregate transactions, but a 2-tier won't)..

                    In most cases where there is a large delta, it has been shown that the "faster" FS has either traded performance for data integrity or have thrown out integrity altogether (there was a KVM/QEMU bug that mean a sync file operation would end up being async. The guest running under KVM was about 100 times faster than the host for the SQLITE test. As it turns out the fsync in the KVM, wasn't really an fsync to the physical disk anyway.
                    Exactly, my intention was not to trash on the benchmark but simply to inform people that there is a huge difference in using SQLite "properly" and not.

                    Comment


                    • #40
                      Originally posted by RussDill View Post
                      Its really too bad more people aren't finding the wiki:

                      http://btrfs.wiki.kernel.org/index.p...sion_from_Ext3
                      Awesome, thanks!

                      Comment

                      Working...
                      X