Linked by Thom Holwerda on Wed 5th Jul 2006 17:09 UTC, submitted by IdaAshley
"The Linux kernel continues to evolve, incorporating new technologies and gaining in reliability, scalability, and performance. One of the most important features of the 2.6 kernel is a scheduler implemented by Ingo Molnar. This scheduler is dynamic, supports load-balancing, and operates in constant time -- O(1). This article explores these attributes of the Linux 2.6 scheduler, and more."
Thread beginning with comment 141021
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
Recent Original Stories
Recent Comments
Headlines
Random Comments
Random Stories
Random OS Link
Member since:
2006-04-17
>Um, no. For the reasons I've already outlined.
No, you did not outline any reasons. A round robin
scheduler will perform badly with IO bound tasks because
it won't let them get in front of the CPU bound tasks.
Sure, the *mechanism* of selecting a task will be very
fast, but it will suck.
>Amazingly enough, I knew that when I made my point.
>Which typing "wrong" is not a refutation of.
Umm, I'll repeat what you wrote:
"any system with more processes than sceduler classes
(ie, nearly any interactive system)"
You were wrong in 1 of 2 ways. Either you thought
the scheduler complexity rose with the total number
of processes in the system (and that's really what
it sounds like), or you thought that "nearly any
interactive system" has a significant number of tasks
on the runqueue at any time.
Here is a sequence of the 'R' field of `vmstat 1` on my
system while browsing the web, listening to music,
editing and compiling:
0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 2 0 0 0 0 3 0 0 1 0 0 5
0 0 0 1 2 2 2 2 3 1 1 ...
You let me know when nearly all your interactive
systems reach 140 or so, and I'll reconsider.