What is a Kernel Overhead?

Question

What is a (micro) kernel overhead and what are the components that increase the cycles in a microkernel? Sources for the information would be more than helpful.

Answer 1

I don't know if you have checked this or not but here it is.

Answer 2

Microkernels strive to make the kernel more secure/fail-safe by removing most OS-level services (drivers, file systems, networking, ...) from privileged kernel mode and putting them into dedicated user-level applications. The microkernel itself then only provides mechanisms to start those applications and for their communication.

The advantage of this approach is that user applications leverage hardware-level isolation (address spaces) and therefore a crashing driver will not take down your whole system. So far the propaganda.

There is an inherent problem though: In a traditional (monolithic) kernel all components run in the same privileged address space and calling e.g., file system functionality is a simple function call. In a microkernel-based system these calls become communication relations through kernel-provided mechanisms, so the cost of such a call rises from 10s of cycles to hundreds of cycles.

This is why microkernel systems are rumored to be slower than traditional operating systems and why microkernel implementors are extremely peculiar about making the kernel-provided communication mechanism as fast as possible.