1. Computers
  2. Display Drivers
  3. Graphics Cards
  4. Memory
  5. Motherboards
  6. Processors
  7. Software
  8. Storage
  9. Operating Systems


Facebook RSS Twitter Twitter Google Plus


Phoronix Test Suite

OpenBenchmarking.org

The Impact Of A Tickless Kernel

Michael Larabel

Published on 24 February 2007
Written by Michael Larabel
Page 1 of 1 - 2 Comments

This will be short and sweet. In this article we will be looking at the impact of CONFIG_NO_HZ/Dynamic Ticks, which will be found in the Linux 2.6.21 kernel. The option has been available as a patch for quite a while, but not until Linux 2.6.21-rc1 had it been merged into the upstream kernel. When enabled, there will only be timer ticks when they are needed. The end-user benefit is cooler-running processors and increased power savings. We have investigated this change with a notebook and desktop computer.

CONFIG_NO_HZ/Dynamic Ticks is supported by SMP systems, but at present the only architecture supported is x86. Support for PowerPC, x86_64, and ARM is under development. For today's testing we had monitored the power consumption of the system using a SeaSonic PowerAngel. We were monitoring the Intel Xeon E5320 temperatures using LM_Sensors, but had found these numbers to be slightly inaccurate which led to their removal from this article. The power was monitored while the system was idling and again during full load caused by compiling the Linux kernel and running glxgears. We had tested the kernel with both CONFIG_NO_HZ (Tickless System (Dynamic Ticks)) and CONFIG_HIGH_RES_TIMERS enabled and then disabled.

The main system we had used was a Tyan Tempest i5000XT motherboard with dual Intel Clovertown E5320 processors and eight sticks of 512MB DDR2 FB-DIMM RAM. We had used this system due to the number of cores and processing power to see how well the tickless kernel would perform. We had also reproduced the same tests using a Lenovo ThinkPad R52 with an Intel Pentium M 750, 2GB of DDR2, and ATI Mobility Radeon X300. On both systems, Fedora 7 Rawhide was used with the Linux 2.6.21-rc1 vanilla kernel.

From these tests we had found the CONFIG_NO_HZ option to cause very little difference on the two test systems used. The Wattage had barely dropped during idle and load with this kernel option. Granted for Intel processors there is already Enhanced Intel SpeedStep Technology (EIST) and Cool 'n' Quiet for AMD processors to reduce the heat output, improve the thermal performance, and drive down your energy bills. However, to reiterate tickless kernels are not brand-specific but at this time it is only compatible with x86 architecture (and also SMP/non-SMP supportive). For those running their systems 24/7 a tickless kernel could end up being an advantage. Likewise, for embedded systems and laptops this ultimately should be of benefit. Outside of the thermal and power advantages, a tickless kernel should improve KVM performance. Our investigating of CONFIG_NO_HZ has just begun and we will report back once we have proceeded with additional tests.

If you enjoyed this article be sure to subscribe to our RSS Feeds and discuss this article in the Phoronix Forums.

About The Author
Michael Larabel is the principal author of Phoronix.com and founded the web-site in 2004 with a focus on enriching the Linux hardware experience and being the largest web-site devoted to Linux hardware reviews, particularly for products relevant to Linux gamers and enthusiasts but also commonly reviewing servers/workstations and embedded Linux devices. Michael has written more than 10,000 articles covering the state of Linux hardware support, Linux performance, graphics hardware drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated testing software. He can be followed via and or contacted via .
Latest Linux Hardware Reviews
  1. Intel Xeon E5-1680 v3 & E5-2687W v3 Compared To The Core i7 5960X On Linux
  2. Intel 120GB 530 Series SSD Linux Performance
  3. Btrfs/EXT4/XFS/F2FS RAID 0/1/5/6/10 Linux Benchmarks On Four SSDs
  4. AMD's Windows Catalyst Driver Remains Largely Faster Than Linux Drivers
Latest Linux Articles
  1. NVIDIA vs. Nouveau Drivers With Linux 3.18 + Mesa 10.4-devel
  2. Is The Open-Source NVIDIA Driver Fast Enough For Steam On Linux Gaming?
  3. Linux 3.18 File-System Performance Minimally Changed But Possible Regressions
  4. AMD Radeon Gallium3D Is Catching Up & Sometimes Beating Catalyst On Linux
Latest Linux News
  1. Gngr: A New Web Browser Focused On Privacy
  2. Linux 3.18 Kernel: Not Much Change With Intel Haswell Performance
  3. More File-System Tests Of The Linux 3.18 Kernel
  4. Using NVIDIA's NVENC On Linux With FFmpeg
  5. There's Talk Again About An "Open To The Core" Ubuntu Laptop
  6. PowerVR SGX Driver Code Gets Leaked
  7. V2 Of KDBUS Published For Linux Kernel Review
  8. VirtualBox 4.3.20 Arrives, Still No Sign Of VirtualBox 4.4
  9. Scientific Linux 6.6 vs. Scientific Linux 7.0 Benchmarks
  10. Qualcomm Looks To Get Into The ARM Server Business
Latest Forum Discussions
  1. Debian Developer Resigns From The Systemd Maintainership Team
  2. Roadmap to Catalyst 14.10 ?
  3. Updated and Optimized Ubuntu Free Graphics Drivers
  4. Cant get working Kaveri APU - A10-7850k
  5. Script for Fan Speed Control
  6. Debian Init System Coupling Vote Results
  7. The Slides Announcing The New "AMDGPU" Kernel Driver
  8. Ubuntu Developers Still Thinking What To Do About Adobe Flash Support