Announcement

Collapse
No announcement yet.

Mobile Users Beware: Linux Has Major Power Regression

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

  • Mobile Users Beware: Linux Has Major Power Regression

    Phoronix: Mobile Users Beware: Linux Has Major Power Regression

    For those that follow my personal Twitter feed will know that for the past week I've been closely testing Ubuntu 11.04 and all Ubuntu releases going back to Ubuntu 8.04 on many mobile devices in the office. The overall system performance, power consumption, and boot performance have been the principal targets. However, late this week I discovered a glaring regression: Ubuntu 11.04 is viciously going through power. Compared to Ubuntu 10.10, the power consumption on Ubuntu 11.04 for mobile devices is up about 10% on average but under some workloads, I am seeing the power consumption up by nearly 30%. This is happening on many mobile systems spanning multiple generations of Intel CPUs and with Intel / ATI / NVIDIA graphics. This issue has been tracked down to a frightening kernel regression in the mainline tree that is still not addressed.

    http://www.phoronix.com/vr.php?view=15926

  • #2
    Blame me, but I'm thinking that latest kernels put the lamp at 100% by default instead older kernels put the lamp at 50% (kidding, but not really seriously) ... how is it possible for such a different hardware to behave all the same way?

    Comment


    • #3
      Originally posted by blackshard View Post
      Blame me, but I'm thinking that latest kernels put the lamp at 100% by default instead older kernels put the lamp at 50% (kidding, but not really seriously) ... how is it possible for such a different hardware to behave all the same way?
      The screen brightness is unchanged.
      Michael Larabel
      http://www.michaellarabel.com/

      Comment


      • #4
        Id really much appreciate an explanation for the X-axis for the Graphs.
        In most cases its time. But it doesnt make sense when you have idling...
        Why are graphs ending at different times if not caused by increased performance?

        Comment


        • #5
          http://xkcd.com/833/

          Comment


          • #6
            Even worse?

            It isn't clear if the power measurements are for the whole system or just the processor. I'm assuming they are for the whole system (ie including screen, drives etc). If that is the case then this problem is even worse. It is reasonable to believe that those physical devices are consuming the same amount of power as before.

            If the devices consumed half the power and the cpu etc consumed the other half then the overall number going up by 10% means all of that increase came from that half of the system. (Sort of the inverse of Amdahl's law.)

            Comment


            • #7
              It boggles the mind how after so many years the rest of the world is still trying to catch up to OS X's power management. You would think in this day and age that it would be common knowledge how to do it properly and not be some black obelisk.

              Comment


              • #8
                Ubuntu is Linux. Linux is not Ubuntu. Just to make that distinction. In my (admittedly limited) experience, Ubuntu is more power hungry by default than many other distros, and I wouldn't use it on a mobile device anyway.
                I noted a distinct lack of anything AMD, or other distros being tested. Is there any particular reason for that? Well Ubuntu only I can guess at, but the lack of AMD hardware has had me intrigued for some time.

                Comment


                • #9
                  Originally posted by mirv View Post
                  Ubuntu is Linux. Linux is not Ubuntu. Just to make that distinction. In my (admittedly limited) experience, Ubuntu is more power hungry by default than many other distros, and I wouldn't use it on a mobile device anyway.
                  I noted a distinct lack of anything AMD, or other distros being tested. Is there any particular reason for that? Well Ubuntu only I can guess at, but the lack of AMD hardware has had me intrigued for some time.
                  Intel as of late sends me lots of hardware where as AMD doesn't, so AMD systems are limited to those that I buy. The last AMD CPUs I received as samples were Opterons a few generations ago. In terms of these mobile tests, I have no mobile AMD devices.
                  Michael Larabel
                  http://www.michaellarabel.com/

                  Comment


                  • #10
                    Originally posted by mirv View Post
                    Ubuntu is Linux.
                    If you are going to clarify do it properly. Ubuntu is a linux distribution, it is not however "linux".

                    Comment


                    • #11
                      Originally posted by deanjo View Post
                      It boggles the mind how after so many years the rest of the world is still trying to catch up to OS X's power management. You would think in this day and age that it would be common knowledge how to do it properly and not be some black obelisk.
                      Only having to deal with specific hardware that you control might have something to do with it

                      Comment


                      • #12
                        Originally posted by deanjo View Post
                        If you are going to clarify do it properly. Ubuntu is a linux distribution, it is not however "linux".
                        If you are going to clarify by clarifying someone else's clarification, do it properly. Ubuntu is a GNU/Linux distribution, it is not (merely) however a "linux distribution".

                        This is in jest, BTW: we could go on forever and list every possible software package:

                        Ubuntu is a GNU/LibreOffice/GNOME/KDE/LXDE/FLTK/FVWM/Unity/Mono/OpenJDK/Firefox/PulseAudio/.................../Linux distribution.

                        Comment


                        • #13
                          Good! Still on 2.6.37 for power consumption (/fan/heat) reasons. So it's not just me.

                          Comment


                          • #14
                            I've been using 2.6.38 for a while now. On my desktop which is running maverick, and and also on my laptop which is running natty so Icould test out unity. I have stuck with 2.6.38 because I have noticed decent latency improvements. While I agree as far a throughput is concerned things don't seem much different, could the various factors improving latency be causing the increase in power usage? Possibly more acpi wakeup events? I have noticed kworker process threads popping up in my conky top 5 cpu usage threads quite frequently.

                            Comment


                            • #15
                              ACPI

                              I think power consumption is not obtained from hardware but calculated by ACPI. It's rather simple mathematics, but it could be the algorithm has been changed in between.
                              If you could try a full battery emptying between, let's say 2.6.37 and 2.6.38 we could be sure about the reported consumption.

                              Comment

                              Working...
                              X