Microkernel VS Monolithic Kernel

This post explains the two main kernel architectures of operating systems: the monolithic kernel and the micro kernel. Starting with an introduction about the term ”kernel” itself and its meaning for operating systems as a whole, it continues with a comparison of benefits and disadvantages of both architectures.

Kernel: kernel is the indispensable and therefore most important part of an operating system. Roughly, an operating system itself consists of two parts: the kernel space (privileged mode) and the user space (unprivileged mode). Without that, protection between the processes would be impossible.

There are two different concepts of kernels:

·         monolithic kernel.

·         μ-kernel (microkernel).

Monolithic kernel: The older approach is the monolithic kernel, of which Unix, MS-DOS and the early Mac OS are typical representants of. It runs every basic system service like process and memory management, interrupt handling and I/O communication, file system, etc. in kernel space see Figure 1 (click here for Anatomy of Linux Kernel). It is constructed in a layered fashion, built up from the fundamental process management up to the interfaces to the rest of the operating system (libraries and on top of them the applications). The inclusion of all basic services in kernel space has three big drawbacks.

·         The kernel size increase.

·         Lack of extensibility.

·         The bad maintainability. 

Figure 1: Monolithic Kernel base Operating System.

Bug-fixing or the addition of new features means a recompilation of the whole kernel. This is time and resource consuming because the compilation of a new kernel can take several hours and a lot of memory. Every time someone adds a new feature or fixes a bug, it means recompilation of the whole

 kernel. (click here: “ How to Compile Kernel Source”)

To overcome these limitations of extensibility and maintain-ability, the idea of μ-kernels appeared at the end of the 1980’s. 

Microkernel: The concept (Figure 2) was to reduce the kernel to basic process communication and I/O control, and let the other system services reside in user space in form of normal processes (as so called servers). There is a server for managing memory issues, one server does process management, another one manages drivers, and so on. Because the servers do not run in kernel space anymore, so called ”con-text switches” are needed, to allow user processes to enter privileged mode (and to exit again). That way, the μ-kernel is not a block of system services anymore, but represents just several basic abstractions and primitives to control the communication between the processes and between a process and the underlying hardware. Because communication is not done in a direct way anymore, a message system is introduced, which allows independent communication and favors extensibility.

Figure 2: Micro Kernel base Operating System.

Currently, there are two different generations of μ-kernels. The first generation was a more or less a stripped-down monolithic kernel. Because of performance drawbacks concerning process communication, several system services like device drivers, communication stacks, etc. found their way back into kernel space. This resulted in an even bigger kernel than before, which was slower than its monolithic counterpart. 

            Research in the field of μ-kernels prove, that it is not the best solution to create a hybrid kernel 1 , but a pure micro kernel, which has to be very small in size. So small, that it fits into the processor’s first level cache as a whole. Second generation μ-kernels like the L4 are highly optimized, not just referring to the processor family, but also to the processor itself 2 , which results in a very good I/O performance.

In a simple way we can say like this :

Monolithic Kernel (Macro Kernel): Kernel Image = (Kernel Core+Kernel Services). When system boots up entire services  are loaded and resides in memory.

Example: Windows and Unix.

Micro kernel : Kernel Image = Kernel Core. Services are build in to special modules which can be loaded and unloaded as per need.

We have another type of kernel integration technique called

Modular, this is derived from best of micro and monolithic kernel) In

Modular kernel integration:  Kernel Image = (Kernel core + IPC service modules +Memory  module +Process Management module). All other modules are loadable kernel modules.

Example: Linux kernel