Linuxlab

Chapter 2

Kernel internals and tuning

Updated July 3, 2026

In Part III you talked to the kernel through /proc, /sys, and sysctl, but you looked at it from the outside: here is a parameter, here is a file, here is a value. Now we work out what you are actually steering. Behind those parameters sit subsystems, large parts of the kernel, each with one job.

Four of them come up most often. The CFS scheduler decides which process gets the CPU right now. The memory manager hands out and reclaims pages, and when no pages are left, the OOM killer steps in. The VFS is a shared layer over every filesystem, so that open and read work the same on ext4 and on NFS. The network stack carries a packet from the network card to a process's socket.

You can touch the scheduler through

Locked

The rest of this chapter is part of a paid course

You are reading the open part. Buy the course to read the whole chapter.

See the course