Announcement

Collapse
No announcement yet.

Ubuntu 10.04 Is Off To A Poor Performance Start

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

  • Ubuntu 10.04 Is Off To A Poor Performance Start

    Phoronix: Ubuntu 10.04 Is Off To A Poor Performance Start

    Tomorrow will mark the first alpha release of Ubuntu 10.04, and while there is still a long journey ahead for this Long-Term Support release before it officially makes its debut in April, we could not pass up the opportunity to provide some early benchmarks of the Lucid Lynx. Ubuntu 10.04 LTS has already pulled in X.Org 7.5 with X Server 1.7 and other updated graphics packages along with the Linux 2.6.32 kernel that it will be using in the final build, which already presents some core differences from the current stable release, Ubuntu 9.10.

    http://www.phoronix.com/vr.php?view=14434

  • #2
    Can someone check to see if Ubuntu uses debugging packages in their pre-releases? Distros used to build them to spot more bugs but they used to run slower

    Comment


    • #3
      At least the notifications use debugging

      Comment


      • #4
        CFQ

        A lot of this benchmarks are server oriented, maybe the patch which make CFQ more responsive for desktop has something to do with that (see http://www.linux-mag.com/id/7637/?ut...ntent=Netvibes)

        Comment


        • #5
          At the moment, lucid has an old xf86-video-ati snapshot which is way slower than current git with KMS (and they're using KMS!). Also, Mesa will hopefully be updated to 7.7.

          Comment


          • #6
            Benchmarking 10.04 alpha 1? Jeez, you guys must be bored..

            How about getting that Win7 build finished so we can finally see those Ubuntu/Mac/Win7 comparisons already??

            Comment


            • #7
              ubuntu debug packages

              I think all packages are debug packages that are then stripped. This extra information is then available in a -dbgsym package. I don't think there are any changes for beta/release time. I know this to be the case as I've submitted crash reports that use these extra packages to give a good stack trace during a crash.

              Comment


              • #8
                Originally posted by jackflap View Post
                Benchmarking 10.04 alpha 1? Jeez, you guys must be bored..
                I'd rather have some info than no info at all, but it is indeed a bit pointless.

                However news always had a function. For example: political climate. If the news pais attention to something than the government will notice it.

                Now let's look at these Aplha 1 benchmarks. It may highlight a problem that Ubuntu devs may come aware of so that they can fix it. We have X.org and Debian people visiting Phoronix. For them this news could be useful so we, as end users, may profit from it on the long run

                Comment


                • #9
                  Originally posted by jackflap View Post
                  Benchmarking 10.04 alpha 1? Jeez, you guys must be bored..

                  How about getting that Win7 build finished so we can finally see those Ubuntu/Mac/Win7 comparisons already??
                  And why not benchmark alpha? It's not like the final release will run 3x faster, the numbers pretty much reflect actual performance that likely will not change.

                  Who is exactly affected by ext4 bug? I'm using "data=writeback,noatime,nodiratime" options.

                  Comment


                  • #10
                    Originally posted by hax0r View Post
                    Who is exactly affected by ext4 bug? I'm using "data=writeback,noatime,nodiratime" options.
                    Wasn't it the 'check a thousand times to see if the file has been written correctly'-'feature' that you could turn of with a flag?

                    And who uses ext4 anyway? XFS is so much better in terms of speed and given the fact that it is journaling makes it enough for me to use anyway...

                    Comment


                    • #11
                      Some precisions

                      Originally posted by hax0r View Post
                      And why not benchmark alpha? It's not like the final release will run 3x faster, the numbers pretty much reflect actual performance that likely will not change.

                      Who is exactly affected by ext4 bug? I'm using "data=writeback,noatime,nodiratime" options.
                      I'm sorry but I totally disagree, the final version can be much faster even more than what you think.

                      I'm studying Software Engineer and when we build a software("debug, alpha" version), they are provided with debug informations and lot's of "bulk" software in it.
                      There's no real reason to build a release version with all the optimizations on it since it's a DEV version.

                      Firstly we do want everything to work as it has to work before compile it as a release version.

                      For example : if you print a single line in console or in a file (it's a very basic instruction but often used "log files etc..") afer every action you do when you ask a SQL database for example. The overall performance of the database will drop down drastically.

                      Of course a test is a mean to have informations on what you are doing.
                      So this bench is not totally useless.

                      But you shouldn't care about drops in performance right now but more about increases
                      The next alpha can have results totally different from this one and the following will follow the same way.
                      Until all optimizations are turned on and all the debug features removed you can't say that's worse.

                      Anyway none of this drops will stay like this, they do take care of performance and all will change but a drop by 90% of performance
                      will not be effective on the release.

                      Hope this help everyone to understand what an "alpha, beta etc." is.
                      Last edited by dl.zerocool; 12-09-2009, 01:08 PM.

                      Comment


                      • #12
                        Originally posted by V!NCENT View Post
                        And who uses ext4 anyway?
                        Anyone with a default Ubuntu install from 9.10 onwards?

                        Comment


                        • #13
                          Originally posted by dl.zerocool View Post
                          I'm sorry but I totally disagree, the final version can be much faster even more than what you think.

                          I'm studying Software Engineer and when we build a software("debug, alpha" version), they are provided with debug informations and lot's of "bulk" software in it.
                          There's no real reason to build a release version with all the optimizations on it since it's a DEV version.

                          Firstly we do want everything to work as it has to work before compile it as a release version.

                          For example : if you print a single line in console or in a file (it's a very basic instruction but often used "log files etc..") afer every action you do when you ask a SQL database for example. The overall performance of the database will drop down drastically.

                          Of course a test is a mean to have informations on what you are doing.
                          So this bench is not totally useless.

                          But you shouldn't care about drops in performance right now but more about increases
                          The next alpha can have results totally different from this one and the following will follow the same way.
                          Until all optimizations are turned on and all the debug features removed you can't say that's worse.

                          Anyway none of this drops will stay like this, they do take care of performance and all will change but a drop by 90% of performance
                          will not be effective on the release.

                          Hope this help everyone to understand what an "alpha, beta etc." is.
                          Ubuntu doesn't follow this development model. It imports packages from debian, it has its own kernel team. It doesn't do a debug/release version of packages, it does stripped debug builds. All the packages that are in ubuntu are not in an alpha state, they're just current versions.

                          Comment


                          • #14
                            Originally posted by movieman View Post
                            Anyone with a default Ubuntu install from 9.10 onwards?
                            You can specify what FS you want to use with Ubuntu 9.10. I did that with Kubuntu 9.10... The option is available from the GUI installer on the LiveCD

                            Comment


                            • #15
                              Originally posted by V!NCENT View Post
                              You can specify what FS you want to use with Ubuntu 9.10. I did that with Kubuntu 9.10... The option is available from the GUI installer on the LiveCD
                              Sure, but most people will just use the defaults; certainly I did when I installed 9.10 on my netbook, though only because there's nothing on there important enough that I'd cry if ext4 bugs ate it .

                              Comment

                              Working...
                              X