CPU utilisation cap?

lukem.freebsd_at_cse.unsw.edu.au
Date: 10/21/04

  • Next message: Lucas Holt: "Re: decreasing interrupt CPU load" 
    To: freebsd-performance@freebsd.org
Date: Thu, 21 Oct 2004 14:23:01 +1000 (EST)

    I have been doing some benchmarking as a part of some driver development
    work, and have encountered a phenomenon I can't explain. I am using
    FreeBSD 5.2.1-RELEASE with SMP and IO-APIC disabled.

    I am using a dual 2.8GHz xeon box, but only one CPU without
    hyperthreading. The box in question has three em interfaces, and one fxp.
    Two of the em's are 133Mhz/64bit, and one is 33MHz/32bit. I have verified
    these values by modifying the em driver to print out what it detects.

    em0: MAC type:82546 r3 Bus speed:133MHz Bus width:64bit Bus type:PCI-X
    em1: MAC type:82546 r3 Bus speed:133MHz Bus width:64bit Bus type:PCI-X
    em2: MAC type:82540 Bus speed:33MHz Bus width:32bit Bus type:PCI

    The particular benchmark I have been using is a UDP echo test, where I
    have a number of linux boxes sending UDP packets to the freebsd box, which
    the freebsd box echoes at user-level (think inetd udp echo, though in fact
    I have also used an optimised server which gets higher throughput).
    Throughput is measured on the boxes which generate the UDP packets.

    I am measuring idle time using a CPU soaker process which runs at a very
    low priority. Top seems to confirm the output it gives.

    What I see is strange. CPU utilisation always peaks (and stays) at between
    80 & 85%. If I increase the amount of work done by the UDP echo program
    (by inserting additional packet copies), CPU utilisation does not rise,
    but rather, throughput declines. The 80% figure is common to both the slow
    and fast PCI cards as well.

    This is rather confusing, as I cannot tell if the system is IO bound or
    CPU bound. Certainly I would not have expected the 133/64 PCI bus to be
    saturated given that peak throughput is around 550Mbit/s with 1024-byte
    packets. (Such a low figure is not unexpected given there are 2 syscalls
    per packet). 

    -- 
Luke
_______________________________________________
freebsd-performance@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "freebsd-performance-unsubscribe@freebsd.org"
  • Next message: Lucas Holt: "Re: decreasing interrupt CPU load"

    Relevant Pages

    • Underutilisation of CPU --- am I PCI bus bandwidth limited?
      ... lukem> satisfactorily answered the question. ... lukem> The particular benchmark I have been using is a UDP echo test, ... lukem> the boxes which generate the UDP packets. ... CPU utilisation always peaks ...
      (freebsd-net)
    • LINUX vs FreeBSD mysql performance using a large RT database
      ... Its not that CPU intensive, but it does get the odd burst of CPU work as mail comes in and is processed by a number of filters. ... With all the threads about poor FreeBSD performance, I wanted to test it out myself to see how 64bit LINUX would compare using the same hardware. ... acpi0: on motherboard ... 0xcc00-0xcc07,0xc880-0xc883,0xc800-0xc807,0xc480-0xc483,0xc400-0xc40f irq 11 at device 14.1 on pci1 ...
      (freebsd-performance)
    • Re: sched_ule performance on single CPU
      ... SCHED_ULE and FreeBSD 7.0 on ... At this very moment I use a private AMD64 box, ... CPU 3500+ at 2,2 ... waiter (in computing, the CPU) to serve. ...
      (freebsd-stable)
    • Re: MySQL Performance 6.0rc1
      ... Nevertheless I run a production MySQL on a double-Xeon (Supermicro ... etc) which eventually kill MySQL with FreeBSD ... CPU: AMD OpteronProcessor 246 ... <ACPI PCI bus> on pcib0 ...
      (freebsd-current)
    • Performance issue amd64
      ... My home's computer system is a FreeBSD 6.1-STABLE/AMD64 box, clean 64 Bit without 32 Bit compat enabled. ... <CPU Frequency Thermal Control> on cpu0 ... <ACPI PCI bus> on pcib0 ... 2 ports with 2 removable, ...
      (freebsd-questions)