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

Ubuntu 10.04 Already Shortens The Boot Time

Michael Larabel

Published on 18 December 2009
Written by Michael Larabel
Page 1 of 1 - 13 Comments

Ubuntu 10.04 Alpha 1 was released last week and while it did not bring any major features yet for this Long-Term Support release of Ubuntu Linux to be released in April of 2010, it began to introduce some underlying changes like the switch to the Linux 2.6.32 kernel, X Server 1.7, and the complete removal of HAL. Our early benchmarks of Ubuntu 10.04 show that there are some negative performance regressions right off the bat, but that is from within the Linux desktop. One area that Canonical is focusing upon in particular with Ubuntu 10.04 LTS is speeding up the boot process, so we decided to provide some benchmarks there too.

For Ubuntu 10.04 in fact they would like a 10 second boot time for Intel netbook hardware. Not only are they speeding up the boot process, but they are also working to beautify the boot process with Plymouth (video). For this article we checked out the boot performance with a Samsung NC10 that had an after-market OCZ SSD and 2GB of RAM along with a stock Dell Mini 9. Starting off, below are the Dell Mini 9 boot chart results for clean installations of Ubuntu 9.10 and then Ubuntu 10.04 Alpha 1.

The reported boot time for Ubuntu 9.10 on the Mini 9 was 59 seconds, but only 24 seconds in comparison for Ubuntu 10.04 Alpha 1. In the past, we had been able to boot Ubuntu 9.10 faster, but these are clean Bootchart results from Ubuntu 9.10 on a stock installation. The peak disk throughput was also higher under Ubuntu Lucid with a speed of 84MB/s compared to 79MB/s with Karmic. With Ubuntu 10.04 Alpha 1 there are less process starting up (such as with HAL being completely nuked) and the processes that are running are running faster. Also, we were measuring the boot chart results for each system and each Ubuntu release with an automated log-in to the GNOME desktop rather than the log-in screen. Below are the charts for the Samsung NC10.

For the Samsung NC10 the Karmic number was 55 seconds compared to 23 seconds with Lucid. The disk throughput for this OCZ Core Series V2 SSD was also noticeably better with Lucid at a peak of 122MB/s compared to 55MB/s with the stable Karmic.

Also surfacing this month is a new kernel (though it is currently based upon Linux 2.6.31 for Ubuntu 9.10) that pulls in more of Intel's Moblin speed patches from their kernel. As most know, Moblin 2 boots extremely fast. As is described on the mailing list, these kernel patches from Moblin remove a time delay in waiting for all devices before mounting the root file-system, KMS after SATA in drivers, don't blank the display, in the kernel DRM removing a root requirement for a DRM ioctl, fixing a vblank delay on pipe disable, make the device initialization go a-synchronously, and add an EDID cash for connected displays. We have tested out this kernel with the Moblin patches on Ubuntu 9.10 and it booted three seconds faster compared to the stock Karmic kernel.

As Ubuntu 10.04 matures, we will be back with more benchmarks. For now though you may be interested in the following articles: Measuring Ubuntu's Boot Performance, Measuring Fedora's Boot Performance, Checking In On Ubuntu Karmic's Boot Time, and ASUS Eee PC 901 Linux Boot Performance.

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. Btrfs On 4 x Intel SSDs In RAID 0/1/5/6/10
  2. AMD Radeon R9 290 On Ubuntu 14.10: RadeonSI Gallium3D vs. Catalyst
  3. MSI X99S SLI PLUS On Linux
  4. NVIDIA GeForce GTX 970 Offers Great Linux Performance
Latest Linux Articles
  1. NVIDIA's Linux Driver Can Deliver Better OpenGL Performance Than Windows 8.1
  2. Windows 8.1 vs. Ubuntu 14.10 With Intel HD Graphics
  3. 6-Way Ubuntu 14.10 Radeon Gallium3D vs. Catalyst Driver Comparison
  4. NVIDIA vs. Nouveau Drivers On Ubuntu 14.10
Latest Linux News
  1. Wine 1.7.30 Continues Work On DirectWrite & Offers Regedit Fixes
  2. Has The Sky Fallen? Qualcomm Contributes To Freedreno's DRM/KMS Driver
  3. Manjaro Works To Make Calamares A Distribution-Independent Installer
  4. DisplayLink USB 3.0 Support Sounds Like A Mess
  5. PulseAudio Gains A Native Bluetooth Headset Backend
  6. X.Org Foundation Decides On Its Women Outreach Project
  7. GTK+ 3.16's New GtkGLArea Widget Gets Improved
  8. X.Org Server 1.17 ABI Bumped
  9. Fedora 21 Beta To Be Released Next Week
  10. Go 1.4 Beta Release Brings Big Runtime Changes
Latest Forum Discussions
  1. Closed source to opensource
  2. What Would You Like To See Next?
  3. How to get rid of Linux
  4. Is foolish currently develop in machine code, hexadecimal and assembly?
  5. Reducing The CPU Usage In Mesa To Improve Performance
  6. Help diagnosing problems with a Readon HD 4670 on Mesa 10.3.2-1
  7. Advertisements On Phoronix
  8. nv and xorg.conf under Debian PPC