Announcement

Collapse
No announcement yet.

Is Fedora 15 Faster Than Ubuntu 11.04?

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

  • Is Fedora 15 Faster Than Ubuntu 11.04?

    Phoronix: Is Fedora 15 Faster Than Ubuntu 11.04?

    Ending out May, we have an interesting set of tests comparing the performance of Fedora 15 and Ubuntu 11.04. This is looking at the "out of the box" performance of both operating systems on three different systems. Not only is the actual test result looked at, but we also compared the power consumption of the two operating systems.

    http://www.phoronix.com/vr.php?view=16077

  • #2
    boot performance???

    it would be interesting i think because of F15 using systemd

    Comment


    • #3
      Well done Michael, I think that's one of the best benchmarking articles I've read on Phoronix ever

      A good set of tests were picked, analysis was done on the data with possible explanations for discrepancies

      For once it didn't feel like a load of graphs dumped out of PTS

      More like this please

      Comment


      • #4
        Originally posted by FireBurn View Post
        Well done Michael, I think that's one of the best benchmarking articles I've read on Phoronix ever

        A good set of tests were picked, analysis was done on the data with possible explanations for discrepancies

        For once it didn't feel like a load of graphs dumped out of PTS

        More like this please
        Thanks... For those liking the articles more than usual in recent days, it's all a matter of May was great regarding some advertising campaigns... For those that try to dismiss ads or complain about them... So just hope more like them come in June, etc.
        Michael Larabel
        http://www.michaellarabel.com/

        Comment


        • #5
          As far as the disk performance discrepancy goes - did you test with SELinux enabled or disabled, on Fedora?

          edit: you say 'out of the box', so I assume 'enabled'; it might be interesting to try with it disabled. Also it'd be interesting to know if you kept the OS default partitioning scheme (in which case the Fedora install would likely have an LVM layer the Ubuntu one would not.)

          Comment


          • #6
            Michael,

            Is PTS still self compiling the tested software?
            If so, have you considered adding, whenever possible the built in version to the mix?

            - Gilboa
            DEV: Intel S2600C0, 2xE52658V2, 32GB, 4x2TB + 2x3TB, GTX780, F21/x86_64, Dell U2711.
            SRV: Intel S5520SC, 2xX5680, 36GB, 4x2TB, GTX550, F21/x86_64, Dell U2412..
            BACK: Tyan Tempest i5400XT, 2xE5335, 8GB, 3x1.5TB, 9800GTX, F21/x86-64.
            LAP: ASUS N56VJ, i7-3630QM, 16GB, 1TB, 635M, F21/x86_64.

            Comment


            • #7
              Where's the conclusion? I didn't really want to read through 6 pages..

              Comment


              • #8
                Originally posted by AdamW View Post
                As far as the disk performance discrepancy goes - did you test with SELinux enabled or disabled, on Fedora?

                edit: you say 'out of the box', so I assume 'enabled'; it might be interesting to try with it disabled. Also it'd be interesting to know if you kept the OS default partitioning scheme (in which case the Fedora install would likely have an LVM layer the Ubuntu one would not.)
                Yes, defaults. Can do some SELinux benchmarks separately if there's interest.
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  Originally posted by DanL View Post
                  Where's the conclusion? I didn't really want to read through 6 pages..
                  Then that detracts from ad views otherwise...
                  Michael Larabel
                  http://www.michaellarabel.com/

                  Comment


                  • #10
                    Originally posted by gilboa View Post
                    Michael,

                    Is PTS still self compiling the tested software?
                    If so, have you considered adding, whenever possible the built in version to the mix?

                    - Gilboa
                    I'd be interested in seeing this too

                    Comparing local version, compiled with local gcc and compiled with phoronix gcc too

                    That would be very interesting

                    Comment


                    • #11
                      fedora 15 & 14 with a hvr2200 can display two windows of tv hd with vlc
                      ubuntu 11 did one tv and was not able to open a second . after that first time vlc was not able to display even 1 tv .
                      from the original install of both os , i copied the file NXP7164-2010-03-10.1.fw in the lib folder and created an options file in the etc/modprobe.d folder with options saa7164 card=4 in it .
                      long ago , ubuntu 9 was perfect with an hvr1100 out of the box with tvtime

                      Comment


                      • #12
                        Originally posted by Michael View Post
                        Yes, defaults. Can do some SELinux benchmarks separately if there's interest.
                        It will be great.

                        Comment


                        • #13
                          The conclusion is [REDACTED]. (just kidding)

                          If you read carefully, the conclusion is on the last page...at the very top, even.

                          Comment


                          • #14
                            Originally posted by AdamW View Post
                            As far as the disk performance discrepancy goes - did you test with SELinux enabled or disabled, on Fedora?

                            edit: you say 'out of the box', so I assume 'enabled'; it might be interesting to try with it disabled. Also it'd be interesting to know if you kept the OS default partitioning scheme (in which case the Fedora install would likely have an LVM layer the Ubuntu one would not.)
                            Good points. It would be interesting to see how much SELinux and LVM affect performance.

                            Comment


                            • #15
                              Originally posted by Michael View Post
                              Yes, defaults. Can do some SELinux benchmarks separately if there's interest.
                              That would be nice. Maybe include AppArmor in those tests as well, although that doesn't use special filesystem labels, but it should decrease application performance as well.

                              Comment

                              Working...
                              X