Re: Disk geommetry, was Re: Kernel Upgrade: Why?

[dlr@postbox.ius.cs.cmu.edu]

When I was installing debian, I made two test partitions on my 12.9GB
IBM "ultra-ATA" drive.  One partition was 300MB from cylinder 6144 to
cylinder 6744, the other was 300MB at the very end of the disk, around
cylinder 25000.  I used the bad block scan on my debian install cd to
check each partition, and timed the whole procedure.  Things took
about twice as long for the partition at cylinder ~25000 as they did
for the partition at cylinder 6144.

I don't know enough to draw conclusions from this, but I did put my
swap partition as near as possible to the "beginning" of my drive.  :)
David LaRose
dlr@cs.cmu.edu

Richard Harran wrote:
> Would I see a performance increase if I made sure my linux partition
> was on the outside of my disk?

Ed wrote:
> Or better yet, some numbers from formal or informal experiments in
> drive partition performance?  

Marsh Ray wrote:
> Yes, I'd like to see this too.  Wouldn't be hard to do, but I don't
> have a spare drive at the moment.