scx/scheds
Andrea Righi c4eb3ce7b4 scx_bpfland: introduce dynamic nvcsw threshold
Instead of using a static value to classify tasks based on their average
amount of voluntary context switches, try to periodically evaluate an
optimal threshold, based on a global average of voluntary context
switches among of all the running tasks.

Tasks with an average amount of voluntary context switches greater than
the global average will be classified as interactive.

The global average is evaluated as an exponentially weighted moving
average (EWMA), as:

  avg(t) = avg(t - 1) * 0.75 - task_avg(t) * 0.25

This approach is more efficient than iterating through all tasks and it
helps to prevent rapid fluctuations that may be caused by bursts of
voluntary context switch events.

The dynamic nvcsw threshold enables a more precise adjustment of
the classification criteria to swiftly respond to global system changes:
tasks can be quickly classified as interactive, but if the system
experiences too many interactive events, the criteria for maintaining
interactive status become stricter. This creates a natural selection
process where only the most deserving tasks remain interactive.

Additionally, introduce the new option `--nvcsw-max-thresh N`, which
allows to extend or restrict the fluctuation range of the global average
threshold for voluntary context switches.

Tested-by: Piotr Gorski <piotrgorski@cachyos.org>
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
2024-07-18 19:03:25 +02:00
..
c Update to vmlinux-v6.10-rc2-g1edab907b57d.h 2024-07-12 11:13:34 -10:00
include Update to vmlinux-v6.10-rc2-g1edab907b57d.h 2024-07-12 11:13:34 -10:00
rust scx_bpfland: introduce dynamic nvcsw threshold 2024-07-18 19:03:25 +02:00
meson.build Restructure scheds folder names 2023-12-17 13:14:31 -08:00
README.md Restructure scheds folder names 2023-12-17 13:14:31 -08:00
sync-to-kernel.sh sync-to-kernel.sh: Sync scx_central and scx_flatcg 2024-02-23 14:21:03 -10:00

SCHED_EXT SCHEDULERS

Introduction

This directory contains the repo's schedulers.

Some of these schedulers are simply examples of different types of schedulers that can be built using sched_ext. They can be loaded and used to schedule on your system, but their primary purpose is to illustrate how various features of sched_ext can be used.

Other schedulers are actually performant, production-ready schedulers. That is, for the correct workload and with the correct tuning, they may be deployed in a production environment with acceptable or possibly even improved performance. Some of the examples could be improved to become production schedulers.

Please see the following README files for details on each of the various types of schedulers:

  • rust describes all of the schedulers with rust user space components. All of these schedulers are production ready.
  • c describes all of the schedulers with C user space components. All of these schedulers are production ready.

Note on syncing

Note that there is a sync-to-kernel.sh script in this directory. This is used to sync any changes to the specific schedulers with the Linux kernel tree. If you've made any changes to a scheduler in please use the script to synchronize with the sched_ext Linux kernel tree:

$ ./sync-to-kernel.sh /path/to/kernel/tree