Speed sensation

[Eray Ozkural]

On Tuesday 22 April 2003 09:21, Roger Larsson wrote:
> If you have a machine with LOTS of RAM (256MB is not enough) you can test
> test this.
> * Time KDE startup from a cold machine.
> * Log out.
> * Time another KDE starup (no restart of computer).
>   This time files are in Linux file cache.

This obviously doesn't test your theory. Any other suggestions?

I am aware of the latency of disk seeks, but as I said the above test doesn't 
measure the number of disk seeks and therefore the additional cost a parallel 
startup would induce compared to potential gains. It was once said that 
parallelized system startup can actually be faster than SysV init scripts so 
similar behavior can possibly be expected in KDE startup (assuming those 
measurements were accurate). There should be a certain degree of parallelism 
(to be found empirically) that would result in improved performance, by about 
15-20% I guess (very wild guess).

Note that this doesn't have anything to do with the fact that the performance 
of shared library loading can be improved by forcing whole files into the 
cache. Since both are sequential file access, maybe it's just lousy disk 
programming in the loader (likely) or a bug in the kernel (again likely)? 
Same gain for all file systems? (not very likely)

Thanks,

-- 
Eray Ozkural (exa) <erayo at cs.bilkent.edu.tr>
Comp. Sci. Dept., Bilkent University, Ankara  KDE Project: http://www.kde.org
www: http://www.cs.bilkent.edu.tr/~erayo  Malfunction: http://mp3.com/ariza
GPG public key fingerprint: 360C 852F 88B0 A745 F31B  EA0F 7C07 AE16 874D 539C