Ask Ubuntu is a question and answer site for Ubuntu users and developers. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I was reading the latest issue of Full Circle (issue 63) and there was an article comparing boot times under various configurations. One of the comparisons was with an installation in a primary partition and with an extended partition. Now, I was shocked to see a graph see significant increases in boot times when the system was installed on an extended partition.

Quickly after, suspicion set in: How is it that simply being in a workaround environment to get around the limitations on the number of partitions allowed magically decreases I/O performance? Is there something I'm missing?

Text from the article:

To my mind, the internal hard drive and the SD card represent just about the respective best and worst physical speeds available for our test computer.

At the same time, I also wanted to know if using a primary hard drive partition (/dev/sda1) or an extended partition (/dev/sda5) had any effect at all. What I got is shown [below].


share|improve this question
up vote 2 down vote accepted

I think that rather than the logical structure, which should not have any impact on performance, you might be observing a difference in speed due to physical locations of the partitions on the drive. The first partition might be faster than the last one.

The key to a fast boot process is low access speed (rather than transfer rates); this is why SSD will allow you to boot in seconds rather than minutes.

However, the access speeds in a hard drive can vary depending on the physical location that the reading heads need to access; the access speeds may vary as much as twofold. The main reason behind that is that as rotation speed is constant for the whole disk, more area passes under the heads on the rim that near the center of the plate. Some more information about that here.

Another thing that might influence the access speed is the partition size: the smaller the partition, the higher the speed. Here you will find the results of a test that shows how the performance can drop depending on how large a partition is.

share|improve this answer
Fascinating, I had never heard of that! Why, specifically, does it do that? Would it be varying strengths of magnetic values on the platter? I thought that nearly all performance was determined by the head and its location/speed of use. – user6658 Oct 9 '12 at 22:04
Nah. It's just the distance of a track to the center of the disk. The rotation speed is same on the rim and at the center, so that the area that passes under a head is larger on the rim. However, I made a blunder in my answer; see correction. – January Oct 9 '12 at 22:17
Thanks for the followup. This explains a lot to me. I appreciate it. – user6658 Oct 10 '12 at 0:09

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.